Re: [RFC PATCH v4 2/4] x86/sgx: Implement support for MADV_WILLNEED

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 2023-06-29 at 17:14 -0700, Sean Christopherson wrote:
> On Thu, Jun 29, 2023, Kai Huang wrote:
> > But SGX usrespace needs a *exact* address.  MAP_FIXED is the only ABI can
> > guarantee this.
> 
> Yes, but you're assuming that userspace is sane and robust.  It's *very* possible
> that there exists userspace that doesn't specify MAP_FIXED, but works anyways
> because their setup doesn't cause conflicts and so the hint is sufficient.

I must have missed something.  ELRANGE is the first VMA that userspace needs to
do, and given enclave address must be within ELRANGE, the subsequent mmap()s
will always conflict with the ELRANGE VMA?  No?




[Index of Archives]     [AMD Graphics]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux