Hi Luis, On Thu, Sep 14, 2017 at 1:39 AM, Luis R. Rodriguez <mcgrof@xxxxxxxxxx> wrote: [snipped] > Would a fw_cache_hint(device, name_list) be reasonable then sometime *before* > suspend? All this would do is ask the firmware API to extend the fw cache > list with the entries. It would not load firmware immediately, instead this > would trigger a request for the hinted firmware to become available for > suspend. Then these drivers could request for firmware at resume and it > will pick up the cached firmware. I think I am the author the patch [1] mentioned by Marcel. I have to admit, it's quite clunky in it's current form. So yes, the new API you proposed is definitely better to solve the issue. I'll send new patch for btusb once we have the new API to use. Also, with patch [1], the "firmware request while host is not available" may still get hit on some corner cases. I proposed another patch [2] to tackle the edge case, can you take a look? [1] https://lkml.org/lkml/2017/8/25/58 [2] https://lkml.org/lkml/2017/8/22/123 Kai-Heng > > Luis > -- > To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
