From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg KH Date: Tue, 08 Jul 2014 17:29:57 +0000 Subject: Re: A question about hotplug and suspend-resume functionallity Message-Id: <20140708172957.GA1797@kroah.com> List-Id: References: <1404838924725.18833@galilsoft.com> In-Reply-To: <1404838924725.18833@galilsoft.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-hotplug@vger.kernel.org On Tue, Jul 08, 2014 at 05:02:57PM +0000, Igor Bezukh wrote: > Hi, > > > We are testing Intel Gigabit adapter driver (igb) under Fedora 20, kernel 3.14.4 for the following use-case: > > (*) Adapter is connected to the PCIE slot > (*) We put the system under suspend by running pm-suspend from user-space > (*) Remove the adapter from the PCIE slot > (*) Wake up the system > > Currently, we got kernel panics and the system gets stuck. Not good, have you sent the kernel panic to the netdev@vger.kernel.org mailing list and cc:ed the driver authors? > My questions are - > 1) in order to support hot-unplugging in igb driver, should I implement the check of device presence in igb_resume > function? > > 2) After the system wakes up, I see in dmesg that the PCI device function igb refuses to change power state and remains in PCI power state D3. > can I assume that after suspend-resume if the function remains in D3 and refuses to move to D1 then hot-unplug event has occured? These questions would be best acked on the above list, the developers there should be able to help you out. good luck, greg k-h