linux-kernel - Re: [PATCH v4 01/15] arm64: rsi: Add RSI definitions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <682d62b8-6cca-4782-b4e4-ffd2a706fadf@redhat.com>
Date: Tue, 23 Jul 2024 15:35:33 +1000
From: Gavin Shan <gshan@...hat.com>
To: Steven Price <steven.price@....com>, kvm@...r.kernel.org,
 kvmarm@...ts.linux.dev
Cc: Suzuki K Poulose <suzuki.poulose@....com>,
 Catalin Marinas <catalin.marinas@....com>, Marc Zyngier <maz@...nel.org>,
 Will Deacon <will@...nel.org>, James Morse <james.morse@....com>,
 Oliver Upton <oliver.upton@...ux.dev>, Zenghui Yu <yuzenghui@...wei.com>,
 linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
 Joey Gouly <joey.gouly@....com>, Alexandru Elisei
 <alexandru.elisei@....com>, Christoffer Dall <christoffer.dall@....com>,
 Fuad Tabba <tabba@...gle.com>, linux-coco@...ts.linux.dev,
 Ganapatrao Kulkarni <gankulkarni@...amperecomputing.com>
Subject: Re: [PATCH v4 01/15] arm64: rsi: Add RSI definitions

On 7/11/24 1:34 AM, Steven Price wrote:
> On 09/07/2024 06:19, Gavin Shan wrote:
>> On 7/1/24 7:54 PM, Steven Price wrote:
>>> From: Suzuki K Poulose <suzuki.poulose@....com>
>>>
>>> The RMM (Realm Management Monitor) provides functionality that can be
>>> accessed by a realm guest through SMC (Realm Services Interface) calls.
>>>
>>> The SMC definitions are based on DEN0137[1] version A-eac5.
>>>
>>> [1] https://developer.arm.com/documentation/den0137/latest
>>>
>>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@....com>
>>> Signed-off-by: Steven Price <steven.price@....com>
>>> ---
>>> Changes since v3:
>>>    * Drop invoke_rsi_fn_smc_with_res() function and call arm_smccc_smc()
>>>      directly instead.
>>>    * Rename header guard in rsi_smc.h to be consistent.
>>> Changes since v2:
>>>    * Rename rsi_get_version() to rsi_request_version()
>>>    * Fix size/alignment of struct realm_config
>>> ---
>>>    arch/arm64/include/asm/rsi_cmds.h |  38 ++++++++
>>>    arch/arm64/include/asm/rsi_smc.h  | 142 ++++++++++++++++++++++++++++++
>>>    2 files changed, 180 insertions(+)
>>>    create mode 100644 arch/arm64/include/asm/rsi_cmds.h
>>>    create mode 100644 arch/arm64/include/asm/rsi_smc.h
>>>
>>
>> [...]
>>
>>> --- /dev/null
>>> +++ b/arch/arm64/include/asm/rsi_smc.h
>>> @@ -0,0 +1,142 @@
>>> +/* SPDX-License-Identifier: GPL-2.0-only */
>>> +/*
>>> + * Copyright (C) 2023 ARM Ltd.
>>> + */
>>> +
>>> +#ifndef __ASM_RSI_SMC_H_
>>> +#define __ASM_RSI_SMC_H_
>>> +
>>> +/*
>>> + * This file describes the Realm Services Interface (RSI) Application
>>> Binary
>>> + * Interface (ABI) for SMC calls made from within the Realm to the
>>> RMM and
>>> + * serviced by the RMM.
>>> + */
>>> +
>>> +#define SMC_RSI_CALL_BASE        0xC4000000
>>> +
>>
>> These fields have been defined in include/linux/arm-smccc.h. Those definitions
>> can be reused. Otherwise, it's not obvious to reader what does 0xC4000000 represent.
>>
>> #define SMC_RSI_CALL_BASE    ((ARM_SMCCC_FAST_CALL << ARM_SMCCC_TYPE_SHIFT)   | \
>>                                   (ARM_SMCCC_SMC_64 << ARM_SMCCC_CALL_CONV_SHIFT) | \
>>                                   (ARM_SMCCC_OWNER_STANDARD << ARM_SMCCC_OWNER_SHIFT))
>>
>> or
>>
>> #define SMC_RSI_CALL_BASE       ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,            \
>>                                                     ARM_SMCCC_SMC_64,               \
>>                                                     ARM_SMCCC_OWNER_STANDARD,       \
>>                                                     0)
> 
> Good point, even better is actually to just drop SMC_RSI_CALL_BASE and
> just redefine SMC_RSI_FID() in terms of ARM_SMCCC_CALL_VAL().
> 

Agreed, it's going to be more clear. The point is to reuse the existing definitions
in include/linux/arm-smccc.h.

Sorry for slow response. I spent some time going through tf-a/tf-rmm implementation
to understand how GPT and stage-2 page-table are managed in order to review this
series. I realized it's complicated to manage GPT and stage-2 page-table and lots
of details still need more time to be figured out.

Thanks,
Gavin