| Message ID | 7d86b2ed20418cab986708e327adb4432f7078c2.camel@gmail.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | usb: gadget: u_ether: Use __netif_rx() in rx_callback() | expand |
On Fri, Sep 06, 2024 at 08:57:25PM +0200, Hubert Wiśniewski wrote: > netif_rx() now disables bottom halves, which causes the USB gadget to be > unable to receive frames if the interface is not brought up quickly enough > after being created by the driver (a bug confirmed on AM3352 SoC). > > Replacing netif_rx() with __netif_rx() restores the old behavior and fixes > the bug. This can be done since rx_callback() is called from the interrupt > context. > > Signed-off-by: Hubert Wiśniewski <wiih@viessmann.com> Doesn't match the "From:" line, so we can't take this :( What commit id does this fix? thanks, greg k-h
On Wed, 2024-09-11 at 15:28 +0200, Greg KH wrote: > On Fri, Sep 06, 2024 at 08:57:25PM +0200, Hubert Wiśniewski wrote: > > netif_rx() now disables bottom halves, which causes the USB gadget to be > > unable to receive frames if the interface is not brought up quickly enough > > after being created by the driver (a bug confirmed on AM3352 SoC). > > > > Replacing netif_rx() with __netif_rx() restores the old behavior and fixes > > the bug. This can be done since rx_callback() is called from the interrupt > > context. > > > > Signed-off-by: Hubert Wiśniewski <wiih@viessmann.com> > > Doesn't match the "From:" line, so we can't take this :( Sorry, my mistake. I will fix that. > What commit id does this fix? This fixes baebdf48c360 ("net: dev: Makes sure netif_rx() can be invoked in any context."). I will add an appropriate Fixes tag in v2. Thank you for your feedback. Hubert Wiśniewski
diff --git a/drivers/usb/gadget/function/u_ether.c b/drivers/usb/gadget/function/u_ether.c index 4bb0553da658..fd7e483b4a48 100644 --- a/drivers/usb/gadget/function/u_ether.c +++ b/drivers/usb/gadget/function/u_ether.c @@ -266,7 +266,7 @@ static void rx_complete(struct usb_ep *ep, struct usb_request *req) /* no buffer copies needed, unless hardware can't * use skb buffers. */ - status = netif_rx(skb2); + status = __netif_rx(skb2); next_frame: skb2 = skb_dequeue(&dev->rx_frames);
netif_rx() now disables bottom halves, which causes the USB gadget to be unable to receive frames if the interface is not brought up quickly enough after being created by the driver (a bug confirmed on AM3352 SoC). Replacing netif_rx() with __netif_rx() restores the old behavior and fixes the bug. This can be done since rx_callback() is called from the interrupt context. Signed-off-by: Hubert Wiśniewski <wiih@viessmann.com> --- drivers/usb/gadget/function/u_ether.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) }