free_channel() has been invoked in
vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
... -> vmbus_close_internal() -> hv_process_channel_removal().
We also change to use list_for_each_entry_safe(), because the entry
is removed in hv_process_channel_removal().
Thank Dan Carpenter for finding the issue!
Signed-off-by: Dexuan Cui <[email protected]>
Reported-by: Dan Carpenter <[email protected]>
Cc: K. Y. Srinivasan <[email protected]>
Cc: Vitaly Kuznetsov <[email protected]>
---
drivers/hv/channel_mgmt.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index 0eeb1b3..865a3af 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -212,11 +212,16 @@ void hv_process_channel_removal(struct vmbus_channel *channel, u32 relid)
void vmbus_free_channels(void)
{
- struct vmbus_channel *channel;
+ struct vmbus_channel *channel, *tmp;
+
+ list_for_each_entry_safe(channel, tmp, &vmbus_connection.chn_list,
+ listentry) {
+ /* if we don't set rescind to true, vmbus_close_internal()
+ * won't invoke hv_process_channel_removal().
+ */
+ channel->rescind = true;
- list_for_each_entry(channel, &vmbus_connection.chn_list, listentry) {
vmbus_device_unregister(channel->device_obj);
- free_channel(channel);
}
}
--
2.1.0
On Wed, Apr 08, 2015 at 11:13:47PM -0700, Dexuan Cui wrote:
> free_channel() has been invoked in
> vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
> ... -> vmbus_close_internal() -> hv_process_channel_removal().
>
> We also change to use list_for_each_entry_safe(), because the entry
> is removed in hv_process_channel_removal().
>
> Thank Dan Carpenter for finding the issue!
>
> Signed-off-by: Dexuan Cui <[email protected]>
> Reported-by: Dan Carpenter <[email protected]>
> Cc: K. Y. Srinivasan <[email protected]>
> Cc: Vitaly Kuznetsov <[email protected]>
> ---
> drivers/hv/channel_mgmt.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
> index 0eeb1b3..865a3af 100644
> --- a/drivers/hv/channel_mgmt.c
> +++ b/drivers/hv/channel_mgmt.c
> @@ -212,11 +212,16 @@ void hv_process_channel_removal(struct vmbus_channel *channel, u32 relid)
>
> void vmbus_free_channels(void)
> {
> - struct vmbus_channel *channel;
> + struct vmbus_channel *channel, *tmp;
> +
> + list_for_each_entry_safe(channel, tmp, &vmbus_connection.chn_list,
> + listentry) {
Horrid indentation, never put continuations on a 1 tab stop location,
you can't read this properly, right?
> + /* if we don't set rescind to true, vmbus_close_internal()
> + * won't invoke hv_process_channel_removal().
> + */
> + channel->rescind = true;
You are changing the logic here, right? Why do you now need the _safe()
call if you aren't freeing anything?
confused,
greg k-h
> -----Original Message-----
> From: Greg KH [mailto:[email protected]]
> Sent: Thursday, April 9, 2015 17:07
> To: Dexuan Cui
> Cc: [email protected]; [email protected];
> [email protected]; [email protected]; [email protected]; KY Srinivasan;
> [email protected]; [email protected]; Haiyang Zhang
> Subject: Re: [PATCH] hv: vmbus_free_channels(): remove the redundant
> free_channel()
>
> On Wed, Apr 08, 2015 at 11:13:47PM -0700, Dexuan Cui wrote:
> > free_channel() has been invoked in
> > vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
> > ... -> vmbus_close_internal() -> hv_process_channel_removal().
> >
> > We also change to use list_for_each_entry_safe(), because the entry
> > is removed in hv_process_channel_removal().
> >
> > Thank Dan Carpenter for finding the issue!
> >
> > Signed-off-by: Dexuan Cui <[email protected]>
> > Reported-by: Dan Carpenter <[email protected]>
> > Cc: K. Y. Srinivasan <[email protected]>
> > Cc: Vitaly Kuznetsov <[email protected]>
> > ---
> > drivers/hv/channel_mgmt.c | 11 ++++++++---
> > 1 file changed, 8 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
> > index 0eeb1b3..865a3af 100644
> > --- a/drivers/hv/channel_mgmt.c
> > +++ b/drivers/hv/channel_mgmt.c
> > @@ -212,11 +212,16 @@ void hv_process_channel_removal(struct
> vmbus_channel *channel, u32 relid)
> >
> > void vmbus_free_channels(void)
> > {
> > - struct vmbus_channel *channel;
> > + struct vmbus_channel *channel, *tmp;
> > +
> > + list_for_each_entry_safe(channel, tmp, &vmbus_connection.chn_list,
> > + listentry) {
>
> Horrid indentation, never put continuations on a 1 tab stop location,
> you can't read this properly, right?
>
> > + /* if we don't set rescind to true, vmbus_close_internal()
> > + * won't invoke hv_process_channel_removal().
> > + */
> > + channel->rescind = true;
>
> You are changing the logic here, right? Why do you now need the _safe()
> call if you aren't freeing anything?
>
> confused,
>
> greg k-h
Hi Greg,
vmbus_device_unregister() invokes device_unregister() -> vmbus_remove() ->
hv_process_channel_removal() -- in this function, we remove the entry from
the list:
list_del(&channel->listentry)
and we run free_channel(channel).
Thanks,
-- Dexuan
On Thu, Apr 09, 2015 at 09:15:30AM +0000, Dexuan Cui wrote:
> > -----Original Message-----
> > From: Greg KH [mailto:[email protected]]
> > Sent: Thursday, April 9, 2015 17:07
> > To: Dexuan Cui
> > Cc: [email protected]; [email protected];
> > [email protected]; [email protected]; [email protected]; KY Srinivasan;
> > [email protected]; [email protected]; Haiyang Zhang
> > Subject: Re: [PATCH] hv: vmbus_free_channels(): remove the redundant
> > free_channel()
> >
> > On Wed, Apr 08, 2015 at 11:13:47PM -0700, Dexuan Cui wrote:
> > > free_channel() has been invoked in
> > > vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
> > > ... -> vmbus_close_internal() -> hv_process_channel_removal().
> > >
> > > We also change to use list_for_each_entry_safe(), because the entry
> > > is removed in hv_process_channel_removal().
> > >
> > > Thank Dan Carpenter for finding the issue!
> > >
> > > Signed-off-by: Dexuan Cui <[email protected]>
> > > Reported-by: Dan Carpenter <[email protected]>
> > > Cc: K. Y. Srinivasan <[email protected]>
> > > Cc: Vitaly Kuznetsov <[email protected]>
> > > ---
> > > drivers/hv/channel_mgmt.c | 11 ++++++++---
> > > 1 file changed, 8 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
> > > index 0eeb1b3..865a3af 100644
> > > --- a/drivers/hv/channel_mgmt.c
> > > +++ b/drivers/hv/channel_mgmt.c
> > > @@ -212,11 +212,16 @@ void hv_process_channel_removal(struct
> > vmbus_channel *channel, u32 relid)
> > >
> > > void vmbus_free_channels(void)
> > > {
> > > - struct vmbus_channel *channel;
> > > + struct vmbus_channel *channel, *tmp;
> > > +
> > > + list_for_each_entry_safe(channel, tmp, &vmbus_connection.chn_list,
> > > + listentry) {
> >
> > Horrid indentation, never put continuations on a 1 tab stop location,
> > you can't read this properly, right?
> >
> > > + /* if we don't set rescind to true, vmbus_close_internal()
> > > + * won't invoke hv_process_channel_removal().
> > > + */
> > > + channel->rescind = true;
> >
> > You are changing the logic here, right? Why do you now need the _safe()
> > call if you aren't freeing anything?
> >
> > confused,
> >
> > greg k-h
>
> Hi Greg,
> vmbus_device_unregister() invokes device_unregister() -> vmbus_remove() ->
> hv_process_channel_removal() -- in this function, we remove the entry from
> the list:
> list_del(&channel->listentry)
> and we run free_channel(channel).
What is "this function" referring to here? vmbus_free_channels()?
still confused.
greg k-h
Thanks,
-- Dexuan
> -----Original Message-----
> From: Greg KH [mailto:[email protected]]
> Sent: Thursday, April 9, 2015 17:22
> To: Dexuan Cui
> Cc: [email protected]; [email protected];
> [email protected]; [email protected]; [email protected]; KY Srinivasan;
> [email protected]; [email protected]; Haiyang Zhang
> Subject: Re: [PATCH] hv: vmbus_free_channels(): remove the redundant
> free_channel()
>
> On Thu, Apr 09, 2015 at 09:15:30AM +0000, Dexuan Cui wrote:
> > > -----Original Message-----
> > > From: Greg KH [mailto:[email protected]]
> > > Sent: Thursday, April 9, 2015 17:07
> > > To: Dexuan Cui
> > > Cc: [email protected]; [email protected];
> > > [email protected]; [email protected]; [email protected]; KY Srinivasan;
> > > [email protected]; [email protected]; Haiyang Zhang
> > > Subject: Re: [PATCH] hv: vmbus_free_channels(): remove the redundant
> > > free_channel()
> > >
> > > On Wed, Apr 08, 2015 at 11:13:47PM -0700, Dexuan Cui wrote:
> > > > free_channel() has been invoked in
> > > > vmbus_remove() -> hv_process_channel_removal(), or vmbus_remove() ->
> > > > ... -> vmbus_close_internal() -> hv_process_channel_removal().
> > > >
> > > > We also change to use list_for_each_entry_safe(), because the entry
> > > > is removed in hv_process_channel_removal().
> > > >
> > > > Thank Dan Carpenter for finding the issue!
> > > >
> > > > Signed-off-by: Dexuan Cui <[email protected]>
> > > > Reported-by: Dan Carpenter <[email protected]>
> > > > Cc: K. Y. Srinivasan <[email protected]>
> > > > Cc: Vitaly Kuznetsov <[email protected]>
> > > > ---
> > > > drivers/hv/channel_mgmt.c | 11 ++++++++---
> > > > 1 file changed, 8 insertions(+), 3 deletions(-)
> > > >
> > > > diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
> > > > index 0eeb1b3..865a3af 100644
> > > > --- a/drivers/hv/channel_mgmt.c
> > > > +++ b/drivers/hv/channel_mgmt.c
> > > > @@ -212,11 +212,16 @@ void hv_process_channel_removal(struct
> > > vmbus_channel *channel, u32 relid)
> > > >
> > > > void vmbus_free_channels(void)
> > > > {
> > > > - struct vmbus_channel *channel;
> > > > + struct vmbus_channel *channel, *tmp;
> > > > +
> > > > + list_for_each_entry_safe(channel, tmp, &vmbus_connection.chn_list,
> > > > + listentry) {
> > >
> > > Horrid indentation, never put continuations on a 1 tab stop location,
> > > you can't read this properly, right?
> > >
> > > > + /* if we don't set rescind to true, vmbus_close_internal()
> > > > + * won't invoke hv_process_channel_removal().
> > > > + */
> > > > + channel->rescind = true;
> > >
> > > You are changing the logic here, right? Why do you now need the _safe()
> > > call if you aren't freeing anything?
> > >
> > > confused,
> > >
> > > greg k-h
> >
> > Hi Greg,
> > vmbus_device_unregister() invokes device_unregister() -> vmbus_remove() ->
> > hv_process_channel_removal() -- in this function, we remove the entry from
> > the list:
> > list_del(&channel->listentry)
> > and we run free_channel(channel).
>
> What is "this function" referring to here? vmbus_free_channels()?
>
> still confused.
>
> greg k-h
"this function" means hv_process_channel_removal().
-- Dexuan