Re: [RFC PATCH 2/4] x86/vdso: x86/sgx: Rework __vdso_sgx_enter_enclave() API

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

 



On Wed, Aug 26, 2020 at 01:20:29PM -0700, Sean Christopherson wrote:
> Andy,
> 
> Any thoughts on using a struct versus a plethora of params?  If you don't
> have a strong opinion, I'd like to push for the struct option as it fixes
> the -EFAULT weirdness, satisfies Nathaniel's request, and gives some
> flexibility for the future.  The code impact, both to the vDSO and to the
> caller, is largely a lateral move, i.e. it's different but in the end
> doesn't require any more or less work.

If we ended up ever considering eBPF idea that I throwed a while ago in
the air, most likely the same struct could be use as the context for the
BPF program and larger portion of the existing vDSO code would remain
intact.

I'm not trying gonzo market BPF.

I'm just pointing out that a struct is a great vessel to travel between
kernel and user space (e.g. a BPF program would be executed by the
kernel by using the vDSO provided context struct).

/Jarkko



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

  Powered by Linux