2012/1/13 Rafał Miłecki [off-list ref]:

I have Asus netbook with BCM4313 since a week, so I'll able to test
this. Just give me a day when I'm back home.

Could we rearrange order of patches to make
bcma: invalidate the mapped window over suspend/resume
this first one and add stable as CC? What do you think about this?

I don't think it is relevant for stable - partly because in stable the
bcma suspend/resume isn't wired up anyway, but mostly because if that
patch really wants to make it into stable, it really is a different
patch than the one I munged it into. The stable bcma doesn't have
bcma_bus_suspend() at all, so it goes into that pci-specific one like
your original email had it, where it's a complete hack.

Does it fix anything in stable?

                     Linus