[PATCH 0/2] Drivers: hv: balloon: Fix some issues in the balloon driver
This patch set adresses a couple of issues discovered in the balloon driver: 1. Prevent stale pressure report being posted. 2. If memory is not brought online within the allowed time, do not roll back hot add state if hot add succeeded. K. Y. Srinivasan (2): Drivers: hv: balloon: Fix a bug in the hot-add code Drivers: hv: balloon: Do not post pressure status if interrupted drivers/hv/hv_balloon.c | 18 ++ 1 files changed, 10 insertions(+), 8 deletions(-) -- 1.7.4.1 ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH 2/2] Drivers: hv: balloon: Do not post pressure status if interrupted
When we are posting pressure status, we may get interrupted and handle the un-balloon operation. In this case just don't post the status as we know the pressure status is stale. Signed-off-by: K. Y. Srinivasan Cc: Stable --- drivers/hv/hv_balloon.c |8 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/drivers/hv/hv_balloon.c b/drivers/hv/hv_balloon.c index bfc7be2..4abaccf 100644 --- a/drivers/hv/hv_balloon.c +++ b/drivers/hv/hv_balloon.c @@ -972,6 +972,14 @@ static void post_status(struct hv_dynmem_device *dm) dm->num_pages_ballooned + compute_balloon_floor(); + /* +* If our transaction ID is no longer current, just don't +* send the status. This can happen if we were interrupted +* after we picked our transaction ID. +*/ + if (status.hdr.trans_id != atomic_read(&trans_id)) + return; + vmbus_sendpacket(dm->dev->channel, &status, sizeof(struct dm_status), (unsigned long)NULL, -- 1.7.4.1 ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH 1/2] Drivers: hv: balloon: Fix a bug in the hot-add code
As we hot-add 128 MB chunks of memory, we wait to ensure that the memory
is onlined before attempting to hot-add the next chunk. If the udev rule for
memory hot-add is not executed within the allowed time, we would rollback the
state and abort further hot-add. Since the hot-add has succeeded and the only
failure is that the memory is not onlined within the allowed time, we should not
be rolling back the state. Fix this bug.
Signed-off-by: K. Y. Srinivasan
Cc: Stable
---
drivers/hv/hv_balloon.c | 10 ++
1 files changed, 2 insertions(+), 8 deletions(-)
diff --git a/drivers/hv/hv_balloon.c b/drivers/hv/hv_balloon.c
index 4c605c7..bfc7be2 100644
--- a/drivers/hv/hv_balloon.c
+++ b/drivers/hv/hv_balloon.c
@@ -562,7 +562,7 @@ static void hv_mem_hot_add(unsigned long start, unsigned
long size,
struct hv_hotadd_state *has)
{
int ret = 0;
- int i, nid, t;
+ int i, nid;
unsigned long start_pfn;
unsigned long processed_pfn;
unsigned long total_pfn = pfn_count;
@@ -608,13 +608,7 @@ static void hv_mem_hot_add(unsigned long start, unsigned
long size,
/*
* Wait for the memory block to be onlined.
*/
- t = wait_for_completion_timeout(&dm_device.ol_waitevent, 5*HZ);
- if (t == 0) {
- pr_info("hot_add memory timedout\n");
- has->ha_end_pfn -= HA_CHUNK;
- has->covered_end_pfn -= processed_pfn;
- break;
- }
+ wait_for_completion_timeout(&dm_device.ol_waitevent, 5*HZ);
}
--
1.7.4.1
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
PRJ007 (tty over tcp driver)
How to take this project up ? The project manager is Saravana Krishnamurthy. How to contact him ? Is this project still alive ? ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
Re: [PATCH 1/2] Drivers: hv: balloon: Fix a bug in the hot-add code
On Fri, Jul 12, 2013 at 06:56:14AM -0700, K. Y. Srinivasan wrote:
> As we hot-add 128 MB chunks of memory, we wait to ensure that the memory
> is onlined before attempting to hot-add the next chunk. If the udev rule for
> memory hot-add is not executed within the allowed time, we would rollback the
> state and abort further hot-add. Since the hot-add has succeeded and the only
> failure is that the memory is not onlined within the allowed time, we should
> not
> be rolling back the state. Fix this bug.
[...]
> /*
>* Wait for the memory block to be onlined.
>*/
> - t = wait_for_completion_timeout(&dm_device.ol_waitevent, 5*HZ);
> - if (t == 0) {
> - pr_info("hot_add memory timedout\n");
> - has->ha_end_pfn -= HA_CHUNK;
> - has->covered_end_pfn -= processed_pfn;
> - break;
> - }
> + wait_for_completion_timeout(&dm_device.ol_waitevent, 5*HZ);
>
> }
>
Well now it might look like a bug that you don't test the result
of wait_for_completion_timeout(). Maybe update the comment to
explain why it's OK to continue anyway?
Ben.
--
Ben Hutchings
We get into the habit of living before acquiring the habit of thinking.
- Albert Camus
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
RE: [PATCH 1/2] Drivers: hv: balloon: Fix a bug in the hot-add code
> -Original Message-
> From: Ben Hutchings [mailto:b...@decadent.org.uk]
> Sent: Friday, July 12, 2013 4:17 PM
> To: KY Srinivasan
> Cc: gre...@linuxfoundation.org; linux-ker...@vger.kernel.org;
> de...@linuxdriverproject.org; o...@aepfle.de; a...@canonical.com;
> jasow...@redhat.com; Stable
> Subject: Re: [PATCH 1/2] Drivers: hv: balloon: Fix a bug in the hot-add code
>
> On Fri, Jul 12, 2013 at 06:56:14AM -0700, K. Y. Srinivasan wrote:
> > As we hot-add 128 MB chunks of memory, we wait to ensure that the memory
> > is onlined before attempting to hot-add the next chunk. If the udev rule for
> > memory hot-add is not executed within the allowed time, we would rollback
> the
> > state and abort further hot-add. Since the hot-add has succeeded and the
> > only
> > failure is that the memory is not onlined within the allowed time, we
> > should not
> > be rolling back the state. Fix this bug.
> [...]
> > /*
> > * Wait for the memory block to be onlined.
> > */
> > - t = wait_for_completion_timeout(&dm_device.ol_waitevent,
> 5*HZ);
> > - if (t == 0) {
> > - pr_info("hot_add memory timedout\n");
> > - has->ha_end_pfn -= HA_CHUNK;
> > - has->covered_end_pfn -= processed_pfn;
> > - break;
> > - }
> > + wait_for_completion_timeout(&dm_device.ol_waitevent,
> 5*HZ);
> >
> > }
> >
>
> Well now it might look like a bug that you don't test the result
> of wait_for_completion_timeout(). Maybe update the comment to
> explain why it's OK to continue anyway?
I put in the comment in the patch explaining why it is ok to continue. To
reiterate,
it is ok to continue because hot add has succeeded. More importantly, what I was
doing earlier - rolling back the state when in fact hot add had succeeded was
incorrect.
Regards,
K. Y
>
> Ben.
>
> --
> Ben Hutchings
> We get into the habit of living before acquiring the habit of thinking.
> - Albert Camus
>
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
Re: [PATCH 1/2] Drivers: hv: balloon: Fix a bug in the hot-add code
On Fri, Jul 12, 2013 at 09:07:19PM +, KY Srinivasan wrote: [...] > > Well now it might look like a bug that you don't test the result > > of wait_for_completion_timeout(). Maybe update the comment to > > explain why it's OK to continue anyway? > > I put in the comment in the patch explaining why it is ok to continue. [...] But that is not nearly as easy to see as the comment that is already *in the code* which your patch isn't updating. Ben. -- Ben Hutchings We get into the habit of living before acquiring the habit of thinking. - Albert Camus ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
