Am Dienstag 15 November 2005 04:19 schrieb jamal: > 1) I think we need to separate the oper state from the rest; so > no need to add dormant to be in netdev_state_t.
ok, it seems that everybody else wants to go with state flags. Even though I'm not convinced, I should accept this and therefore Thomas' patch should be the base for continued work. I can concentrate on userspace dormant interaction as I use WPA everyday and know people that can test 802.1X. > 2) Events need only be generated from/to down state At least for ethernet devices, the userspace supplicant needs to know when a down->dormant transition happens, and routing daemons/dhcp needs to know about a transition to up, we cannot filter events. > 3) IFF_WAIT is not needed. A device goes from NOTPRESENT > to DOWN; and may go to DORMANT if there is a protocol negotiation > needed. In other words, DORMANT is a "healthy" state really. I already commented on that in another mail, we need IFF_WAIT. > On the positive side: > > I think the IFF_XXX flags that you have are the way to go. > Stefan's approach to dev->operstate_kernel is the way to go for kernel > side. But at this point I don't get it. In your response to my patch, you say: > b) I think we need the dev->operstate_kernel to behave in a similar > manner as dev->state. Despite that we already have rtnl_lock at that > point it will be consistent to use the set/clear_bit etc. Now what? Is a state field or are state flags your favorite way to go inside the kernel? Would you prefer seperate bit fields for operstate both inside the kernel and in userspace communication so that they don't clutter net_device->flags and net_device->state? Please make up your mind. > I think i have comments on the state transitions, but it will be more > appropriate to make them when theres one patch existing. No. If you have comments on state transitions, make them now because they will affect the netif_*_on/off() functions we need to create when going the bitfield way. Stefan - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
