Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 16 Jun 2021 19:11:42 +1000
From:   Nicholas Piggin <npiggin@gmail.com>
Subject: Re: [PATCH v5 04/17] powerpc/vas: Add platform specific user window
 operations
To:     Haren Myneni <haren@linux.ibm.com>, herbert@gondor.apana.org.au,
        linux-crypto@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
        mpe@ellerman.id.au
Cc:     haren@us.ibm.com, hbabu@us.ibm.com
References: <ed7a09822cf3a2e463f942e5a37309a2365c9d79.camel@linux.ibm.com>
        <c64fda6e9b684c175cedb3ec448cce7aaf0f4615.camel@linux.ibm.com>
        <1623636838.8tsdy9nisc.astroid@bobo.none>
        <e409750a3bd5f8410d7a8a290c69375486420b93.camel@linux.ibm.com>
In-Reply-To: <e409750a3bd5f8410d7a8a290c69375486420b93.camel@linux.ibm.com>
MIME-Version: 1.0
Message-Id: <1623834468.xkn6to1dom.astroid@bobo.none>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Excerpts from Haren Myneni's message of June 15, 2021 4:37 pm:
> On Mon, 2021-06-14 at 12:24 +1000, Nicholas Piggin wrote:
>> Excerpts from Haren Myneni's message of June 13, 2021 8:57 pm:
>> > PowerNV uses registers to open/close VAS windows, and getting the
>> > paste address. Whereas the hypervisor calls are used on PowerVM.
>> >=20
>> > This patch adds the platform specific user space window operations
>> > and register with the common VAS user space interface.
>> >=20
>> > Signed-off-by: Haren Myneni <haren@linux.ibm.com>
>> > ---
>> >  arch/powerpc/include/asm/vas.h              | 14 +++++-
>> >  arch/powerpc/platforms/book3s/vas-api.c     | 53 +++++++++++++--
>> > ------
>> >  arch/powerpc/platforms/powernv/vas-window.c | 45 ++++++++++++++++-
>> >  3 files changed, 89 insertions(+), 23 deletions(-)
>> >=20
>> > diff --git a/arch/powerpc/include/asm/vas.h
>> > b/arch/powerpc/include/asm/vas.h
>> > index bab7891d43f5..85318d7446c7 100644
>> > --- a/arch/powerpc/include/asm/vas.h
>> > +++ b/arch/powerpc/include/asm/vas.h
>> > @@ -5,6 +5,7 @@
>> > =20
>> >  #ifndef _ASM_POWERPC_VAS_H
>> >  #define _ASM_POWERPC_VAS_H
>> > +#include <uapi/asm/vas-api.h>
>> > =20
>> >  struct vas_window;
>> > =20
>> > @@ -48,6 +49,16 @@ enum vas_cop_type {
>> >  	VAS_COP_TYPE_MAX,
>> >  };
>> > =20
>> > +/*
>> > + * User space window operations used for powernv and powerVM
>> > + */
>> > +struct vas_user_win_ops {
>> > +	struct vas_window * (*open_win)(struct vas_tx_win_open_attr *,
>> > +				enum vas_cop_type);
>> > +	u64 (*paste_addr)(struct vas_window *);
>> > +	int (*close_win)(struct vas_window *);
>> > +};
>>=20
>> This looks better, but rather than pull in uapi and the user API=20
>> structure here, could you just pass in vas_id and flags after the
>> common=20
>> code does the user copy and verifies the version and other details?
>>=20
>> I think it's generally good practice to limit the data that the usre
>> can influence as much as possible. Sorry for not picking up on that
>> earlier.
>=20
> The user space pass vas_tx_win_open_attr struct - use only vas_id and
> flags right now but it can be extended in future with reserve elements.
> So passing the same struct to platform specific API.
>=20
> do you prefer "struct vas_window * (*open_win)(vas_id, flags, cop)" and
> extend later when more elments are used?

Yes I think so. The reason being so you don't sending data under the
control of user very far into the kernel. Better safe than sorry.

Thanks,
Nick