Issue with sched-no-work-when-pi-blocked.patch
Hi,
on a system running 3.2.23-rt37 the boot time is increased by two minutes because of commit 4b03ee27 (sched-no-work-when-pi-blocked.patch) when booting with ip=dhcp. This is due to ic_open_devs taking rtnl_mutex before it ups the network interfaces. The ndo_open function then calls netif_carrier_off which queues a work that as well takes rtnl_mutex. This stalls the workqueue (tsk_is_pi_blocked()) and prevents phy_state_machine and thus netif_carrier_on from being run. ic_open_devs waits for two minutes for a link before it gives up and releases rtnl_mutex.
So, what is the rationale behind sched-no-work-when-pi-blocked.patch?
On the other hand, I think I'll just drop rtnl_mutex in ic_open_devs while it is waiting for a link.
Daniel
Signature · 11 lines
Dipl.-Math. Daniel Glöckner, emlix GmbH, http://www.emlix.com Fon +49 551 30664-0, Fax +49 551 30664-11, Bertha-von-Suttner-Straße 9, 37085 Göttingen, Germany Sitz der Gesellschaft: Göttingen, Amtsgericht Göttingen HR B 3160 Geschäftsführer: Dr. Uwe Kracke, Ust-IdNr.: DE 205 198 055 emlix - your embedded linux partner -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html