2008-02-04 13:05:59

by Jan-Bernd Themann

[permalink] [raw]
Subject: [PATCH 2/2] ehea: add memory remove hotplug support

Add memory remove hotplug support

Signed-off-by: Jan-Bernd Themann <[email protected]>

---
Comment: This patch depends on the following patch that
exports the symbols
register_memory_notifier()
unregister_memory_notifier()

http://lkml.org/lkml/2008/2/1/293


drivers/net/ehea/ehea_main.c | 25 +++++++++++++++++++++++++
1 files changed, 25 insertions(+), 0 deletions(-)

diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index 21af674..b75afcc 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -35,6 +35,7 @@
#include <linux/if_ether.h>
#include <linux/notifier.h>
#include <linux/reboot.h>
+#include <linux/memory.h>
#include <asm/kexec.h>

#include <net/ip.h>
@@ -3480,6 +3481,24 @@ void ehea_crash_handler(void)
0, H_DEREG_BCMC);
}

+static int ehea_mem_notifier(struct notifier_block *nb,
+ unsigned long action, void *data)
+{
+ switch (action) {
+ case MEM_OFFLINE:
+ ehea_info("memory has been removed");
+ ehea_rereg_mrs(NULL);
+ break;
+ default:
+ break;
+ }
+ return NOTIFY_OK;
+}
+
+static struct notifier_block ehea_mem_nb = {
+ .notifier_call = ehea_mem_notifier,
+};
+
static int ehea_reboot_notifier(struct notifier_block *nb,
unsigned long action, void *unused)
{
@@ -3559,6 +3578,10 @@ int __init ehea_module_init(void)
if (ret)
ehea_info("failed registering reboot notifier");

+ ret = register_memory_notifier(&ehea_mem_nb);
+ if (ret)
+ ehea_info("failed registering memory remove notifier");
+
ret = crash_shutdown_register(&ehea_crash_handler);
if (ret)
ehea_info("failed registering crash handler");
@@ -3582,6 +3605,7 @@ int __init ehea_module_init(void)
out3:
ibmebus_unregister_driver(&ehea_driver);
out2:
+ unregister_memory_notifier(&ehea_mem_nb);
unregister_reboot_notifier(&ehea_reboot_nb);
crash_shutdown_unregister(&ehea_crash_handler);
out:
@@ -3599,6 +3623,7 @@ static void __exit ehea_module_exit(void)
ret = crash_shutdown_unregister(&ehea_crash_handler);
if (ret)
ehea_info("failed unregistering crash handler");
+ unregister_memory_notifier(&ehea_mem_nb);
kfree(ehea_fw_handles.arr);
kfree(ehea_bcmc_regs.arr);
ehea_destroy_busmap();
--
1.5.2


2008-02-04 14:46:23

by Michael Ellerman

[permalink] [raw]
Subject: Re: [PATCH 2/2] ehea: add memory remove hotplug support

On Mon, 2008-02-04 at 14:04 +0100, Jan-Bernd Themann wrote:
> Add memory remove hotplug support
>
> Signed-off-by: Jan-Bernd Themann <[email protected]>
>
> ---
> Comment: This patch depends on the following patch that
> exports the symbols
> register_memory_notifier()
> unregister_memory_notifier()
>
> http://lkml.org/lkml/2008/2/1/293
>
>
> drivers/net/ehea/ehea_main.c | 25 +++++++++++++++++++++++++
> 1 files changed, 25 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
> index 21af674..b75afcc 100644
> --- a/drivers/net/ehea/ehea_main.c
> +++ b/drivers/net/ehea/ehea_main.c

> @@ -3559,6 +3578,10 @@ int __init ehea_module_init(void)
> if (ret)
> ehea_info("failed registering reboot notifier");
>
> + ret = register_memory_notifier(&ehea_mem_nb);
> + if (ret)
> + ehea_info("failed registering memory remove notifier");
>
> ret = crash_shutdown_register(&ehea_crash_handler);
> if (ret)
> ehea_info("failed registering crash handler");

You don't do anything except print a message if the registration fails.
What happens when someone tries to remove memory but the memory notifier
wasn't registered properly? Bang?

cheers

--
Michael Ellerman
OzLabs, IBM Australia Development Lab

wwweb: http://michael.ellerman.id.au
phone: +61 2 6212 1183 (tie line 70 21183)

We do not inherit the earth from our ancestors,
we borrow it from our children. - S.M.A.R.T Person


Attachments:
signature.asc (189.00 B)
This is a digitally signed message part

2008-02-04 15:24:50

by Jan-Bernd Themann

[permalink] [raw]
Subject: Re: [PATCH 2/2] ehea: add memory remove hotplug support

On Monday 04 February 2008 15:46, Michael Ellerman wrote:
> On Mon, 2008-02-04 at 14:04 +0100, Jan-Bernd Themann wrote:
> > Add memory remove hotplug support

> > @@ -3559,6 +3578,10 @@ int __init ehea_module_init(void)
> > if (ret)
> > ehea_info("failed registering reboot notifier");
> >
> > + ret = register_memory_notifier(&ehea_mem_nb);
> > + if (ret)
> > + ehea_info("failed registering memory remove notifier");
> >
> > ret = crash_shutdown_register(&ehea_crash_handler);
> > if (ret)
> > ehea_info("failed registering crash handler");
>
> You don't do anything except print a message if the registration fails.
> What happens when someone tries to remove memory but the memory notifier
> wasn't registered properly? Bang?

In case the registration fails and somebody tries to free memory:
- Driver will not remove the affected memory from the eHEA memory region
--> Firmware (phyp) can not free that memory (as marked as used)
--> Therefore the removed memory could not be used in an other partition

It makes sense to allow the driver to work anyway. Having no ethernet
would not really be a good alternative.

Regards,
Jan-Bernd

2008-02-04 23:14:52

by Michael Ellerman

[permalink] [raw]
Subject: Re: [PATCH 2/2] ehea: add memory remove hotplug support


On Mon, 2008-02-04 at 16:24 +0100, Jan-Bernd Themann wrote:
> On Monday 04 February 2008 15:46, Michael Ellerman wrote:
> > On Mon, 2008-02-04 at 14:04 +0100, Jan-Bernd Themann wrote:
> > > Add memory remove hotplug support
>
> > > @@ -3559,6 +3578,10 @@ int __init ehea_module_init(void)
> > > if (ret)
> > > ehea_info("failed registering reboot notifier");
> > >
> > > + ret = register_memory_notifier(&ehea_mem_nb);
> > > + if (ret)
> > > + ehea_info("failed registering memory remove notifier");
> > >
> > > ret = crash_shutdown_register(&ehea_crash_handler);
> > > if (ret)
> > > ehea_info("failed registering crash handler");
> >
> > You don't do anything except print a message if the registration fails.
> > What happens when someone tries to remove memory but the memory notifier
> > wasn't registered properly? Bang?
>
> In case the registration fails and somebody tries to free memory:
> - Driver will not remove the affected memory from the eHEA memory region
> --> Firmware (phyp) can not free that memory (as marked as used)
> --> Therefore the removed memory could not be used in an other partition
>
> It makes sense to allow the driver to work anyway. Having no ethernet
> would not really be a good alternative.

Yeah I agree. I wasn't clear from the patch that no notifier -> no
memory removed. Rather than, no notifier -> memory removed and ehea
isn't told about it.

cheers

--
Michael Ellerman
OzLabs, IBM Australia Development Lab

wwweb: http://michael.ellerman.id.au
phone: +61 2 6212 1183 (tie line 70 21183)

We do not inherit the earth from our ancestors,
we borrow it from our children. - S.M.A.R.T Person


Attachments:
signature.asc (189.00 B)
This is a digitally signed message part