Messages in this thread | | | Date | Thu, 13 Mar 2008 21:28:20 +0100 | From | Haavard Skinnemoen <> | Subject | Re: [PATCH 07/10] macb: Basic suspend/resume support |
| |
On Thu, 13 Mar 2008 12:18:56 -0800 David Brownell <david-b@pacbell.net> wrote:
> On Thursday 13 March 2008, Haavard Skinnemoen wrote: > > --- > > drivers/net/macb.c | 37 +++++++++++++++++++++++++++++++++++++ > > 1 files changed, 37 insertions(+), 0 deletions(-) > > Shouldn't this look more like the suspend/resume support > in drivers/net/arm/at91_ether.c ? I don't see calls to > netif_stop_queue()/netif_start_queue(), among other things.
Could be, although netif_stop_queue()/netif_start_queue() is done by netif_device_detach() and netif_device_attach() respectively.
We should probably shut things down gracefully as well...not nice to disable the clock when the MAC is in the middle of transmitting a frame. Like I said in the introductory mail, this patch is just a quick hack to prevent the macb from waking the system prematurely.
> The fact that this driver leaves the peripheral clocked > when it's not open is a minor PM goof; ditto not using > the PHY irq to track connect/disconnect. But that's just > general PM stuff, not specific to suspend/resume support.
Right, it's on my todo list.
Haavard
| |