Re: Sealed memfd & no-fault mmap

[Peter Xu]

On Thu, Apr 29, 2021 at 06:48:07PM +0300, Kirill A. Shutemov wrote:
> > Side note: this will only ever work for non-shared mappings.
> 
> I think it's show-stopper for the use-case, no? IIUC, the mappings is used
> for communication between a compositor and a client and has to be shared.

Yes I had the same doubt..  Besides, we probably don't want to convert all
VM_FAULT_SIGBUS to fallback to anonymous pages, as I see that vmf_error()
converts mostly everything besides -ENOMEM to VM_FAULT_SIGBUS.  Thanks,

-- 
Peter Xu