On 6/16/21 11:35 AM, Christoph Hellwig wrote: > On Wed, Jun 16, 2021 at 11:33:50AM -0400, Boris Ostrovsky wrote: >> Isn't the expectation of virt_to_page() that it only works on non-vmalloc'd >> addresses? (This is not a rhetorical question, I actually don't know). > Yes. Thus is why I'd suggest to just do the vmalloc_to_page or > virt_to_page dance in ops_helpers.c and just continue using that.
Ah, OK, so something along the lines of what I suggested. (I thought by "helpers" you meant virt_to_page()). -boris
