On Fri, 16 Jan 2015, Rajat Jain wrote:
2b3940b60626 ("PCI: pciehp: Remove a non-existent card, regardless of "surprise" capability") made it so we handle unexpected loss of presence detect even if the slot doesn't advertise Hot-Plug Surprise. I suppose we could make a similar argument that if a card shows up, maybe we should do something with it regardless of what the slot advertises for Hot-Plug Capable.Ummm ... may be not. From The PCI Hotplug Spec ver 1.1. section 2.2.2 (Hot Insertion) ================================================================== A slot must be powered down and isolated from the bus before an add-in card can be inserted. The process of making a slot ready for insertion depends upon the particular Platform and operating system. The following general sequence of steps is necessary to insert an add-in card into a slot after it is powered down and ready for insertion: 1. The user inserts the new add-in card. 2. The user notifies the Hot-Plug Service to turn on the slot containing the new add-in card. 3. The Hot-Plug Service issues a Hot-Plug Primitive to the Hot-Plug System Driver to turn on the appropriate slot. =============================================================== Also, I recall I read somewhere in the specs about the need to use the button to initiate hot-plug. This change would make the button useless, no? I'm thinking the user may have reasons to have the card plugged-in, but not have it turned on. The card may be a back-up for example, to be turned on (to enable an alternate path) in case of some other card fails etc. Or may be other application reasons. Thoughts?
Maybe I need to try this on an add-in card. I'm using front-loading SFF devices which don't have attention buttons. I would have expected these slots to have the hot-plug surprise capability set, but there's enough platforms in existnace that don't have it set, but can handle add events just fine otherwise. So while this patch fixes that scenario, I don't want to break others. -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
