RE: [PATCH v19,RESEND 16/27] x86/sgx: Add the Linux SGX Enclave Driver

[Huang, Kai]

> On Tue, Mar 26, 2019 at 01:40:57PM +0100, Thomas Gleixner wrote:
> > On Tue, 26 Mar 2019, Huang, Kai wrote:
> > > On Wed, 2019-03-20 at 18:21 +0200, Jarkko Sakkinen wrote:
> > > >  13 files changed, 1657 insertions(+), 2 deletions(-)  create mode
> > > > 100644 arch/x86/include/uapi/asm/sgx.h  create mode 100644
> > > > arch/x86/kernel/cpu/sgx/driver/Makefile
> > > >  create mode 100644 arch/x86/kernel/cpu/sgx/driver/driver.h
> > > >  create mode 100644 arch/x86/kernel/cpu/sgx/driver/ioctl.c
> > > >  create mode 100644 arch/x86/kernel/cpu/sgx/driver/main.c
> > > >  create mode 100644 arch/x86/kernel/cpu/sgx/encl.c  create mode
> > > > 100644 arch/x86/kernel/cpu/sgx/encl.h
> > >
> > > Shouldn't the driver be located somewhere under drivers/, but not under
> arch/x86?
> > >
> > > I don't think x86 maintainers should have the burden to review every
> > > code change made to SGX driver?
> >
> > I rather prefer to have it in x86. Why? Because driver code has a
> > tendency to get under the radar.
> 
> And having everything under arch/x86 will likely reduce the maintenance
> burden for everyone:
> 
>    - Doesn't require taking changes through multiple trees or coordinating
>      acks from multiple maintainers.
> 
>    - Significantly reduces the number of functions, macros, structs and
>      variables that needs to be exposed in asm/sgx.h (actually eliminates
>      it entirely at this point) which allows for sane code organization as
>      opposed to dumping everything in one big header.
> 
>    - Mostly avoids bikeshedding over whether something is architectural
>      or belongs in the so called driver.

Sure. Thanks.

Thanks,
-Kai