2018-04-10 14:39:41

by Jia-Ju Bai

Subject: [PATCH] net: usb: hso: Replace GFP_ATOMIC with GFP_KERNEL in hso_create_device

hso_create_device() is never called in atomic context.

The call chains ending up at hso_create_device() are:
[1] hso_create_device() <- hso_create_bulk_serial_device() <- hso_probe()
[2] hso_create_device() <- hso_create_mux_serial_device() <- hso_probe()
[3] hso_create_device() <- hso_create_net_device() <- hso_probe()
hso_probe() is set as ".probe" in struct usb_driver,
so it is not called in atomic context.

Despite never getting called from atomic context,
hso_create_device() calls kzalloc() with GFP_ATOMIC,
which does not sleep for allocation.
GFP_ATOMIC is not necessary and can be replaced with GFP_KERNEL,
which can sleep and improve the possibility of sucessful allocation.

This is found by a static analysis tool named DCNS written by myself.
And I also manually check it.

Signed-off-by: Jia-Ju Bai <[email protected]>
---
drivers/net/usb/hso.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c
index d7a3379..3d7a33f 100644
--- a/drivers/net/usb/hso.c
+++ b/drivers/net/usb/hso.c
@@ -2332,7 +2332,7 @@ static struct hso_device *hso_create_device(struct usb_interface *intf,
{
struct hso_device *hso_dev;

- hso_dev = kzalloc(sizeof(*hso_dev), GFP_ATOMIC);
+ hso_dev = kzalloc(sizeof(*hso_dev), GFP_KERNEL);
if (!hso_dev)
return NULL;

--
1.9.1

2018-04-10 15:07:02

by Johan Hovold

Subject: Re: [PATCH] net: usb: hso: Replace GFP_ATOMIC with GFP_KERNEL in hso_create_device

On Tue, Apr 10, 2018 at 10:35:24PM +0800, Jia-Ju Bai wrote:
> hso_create_device() is never called in atomic context.
>
> The call chains ending up at hso_create_device() are:
> [1] hso_create_device() <- hso_create_bulk_serial_device() <- hso_probe()
> [2] hso_create_device() <- hso_create_mux_serial_device() <- hso_probe()
> [3] hso_create_device() <- hso_create_net_device() <- hso_probe()
> hso_probe() is set as ".probe" in struct usb_driver,
> so it is not called in atomic context.
>
> Despite never getting called from atomic context,
> hso_create_device() calls kzalloc() with GFP_ATOMIC,
> which does not sleep for allocation.
> GFP_ATOMIC is not necessary and can be replaced with GFP_KERNEL,
> which can sleep and improve the possibility of sucessful allocation.
>
> This is found by a static analysis tool named DCNS written by myself.
> And I also manually check it.
>
> Signed-off-by: Jia-Ju Bai <[email protected]>

Thanks for the patch. This looks good to me.

Reviewed-by: Johan Hovold <[email protected]>

Johan