Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH 3/3] MIPS: cpu-probe: Vulnerabilities for Loongson cores
To:     WANG Xuerui <i@xen0n.name>, linux-mips@vger.kernel.org
Cc:     Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
        Alexey Malahov <Alexey.Malahov@baikalelectronics.ru>,
        Serge Semin <Sergey.Semin@baikalelectronics.ru>,
        =?UTF-8?B?5ZGo55Cw5p2wIChaaG91IFlhbmppZSk=?= 
        <zhouyanjie@wanyeetech.com>, Paul Cercueil <paul@crapouillou.net>,
        Tiezhu Yang <yangtiezhu@loongson.cn>,
        Huacai Chen <chenhc@lemote.com>, YunQiang Su <syq@debian.org>,
        Liangliang Huang <huanglllzu@gmail.com>,
        linux-kernel@vger.kernel.org
References: <20201230032314.10042-1-jiaxun.yang@flygoat.com>
 <20201230032314.10042-3-jiaxun.yang@flygoat.com>
 <d9764581-249f-166d-966b-c1b2997a27a7@xen0n.name>
From:   Jiaxun Yang <jiaxun.yang@flygoat.com>
Message-ID: <81981263-da62-569f-1b47-b258aa5c91e5@flygoat.com>
Date:   Fri, 1 Jan 2021 00:09:35 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101
 Thunderbird/78.6.0
MIME-Version: 1.0
In-Reply-To: <d9764581-249f-166d-966b-c1b2997a27a7@xen0n.name>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


在 2020/12/31 23:43, WANG Xuerui 写道:
> Hi Jiaxun,
>
> On 12/30/20 11:23 AM, Jiaxun Yang wrote:
>> Loongson64C is known to be vulnerable to meltdown according to
>> PoC from Rui Wang <r@hev.cc>.
>>
>> Loongson64G defended these side-channel attack by silicon.
> "Loongson64G mitigated it in hardware"?
>>
>> Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
>> ---
>>   arch/mips/kernel/cpu-probe.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>
>> diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
>> index 2460783dbdb1..24b21f51353c 100644
>> --- a/arch/mips/kernel/cpu-probe.c
>> +++ b/arch/mips/kernel/cpu-probe.c
>> @@ -2092,6 +2092,8 @@ static inline void cpu_probe_loongson(struct 
>> cpuinfo_mips *c, unsigned int cpu)
>>           c->ases |= (MIPS_ASE_LOONGSON_MMI | MIPS_ASE_LOONGSON_CAM |
>>               MIPS_ASE_LOONGSON_EXT | MIPS_ASE_LOONGSON_EXT2);
>>           c->ases &= ~MIPS_ASE_VZ; /* VZ of Loongson-3A2000/3000 is 
>> incomplete */
>> +        c->vulnerabilities |= MIPS_VULNBL_MELTDOWN;
>> +        c->vulnerable |= MIPS_VULNBL_MELTDOWN;
>>           break;
>>       case PRID_IMP_LOONGSON_64G:
>>           c->cputype = CPU_LOONGSON64;
>> @@ -2100,6 +2102,8 @@ static inline void cpu_probe_loongson(struct 
>> cpuinfo_mips *c, unsigned int cpu)
>>           set_isa(c, MIPS_CPU_ISA_M64R2);
>>           decode_cpucfg(c);
>>           c->writecombine = _CACHE_UNCACHED_ACCELERATED;
>> +        c->vulnerabilities |= MIPS_VULNBL_MELTDOWN |
>> +                  MIPS_VULNBL_SPECTRE_V1 | MIPS_VULNBL_SPECTRE_V2;
>
> Of course you forgot to set the "mitigated" mask... Oh wait.

Hi Xuerui,

Actually it belongs to not affected category as there is no action
to take in kernel.

>
> It seems the "mitigated" mask in the 1st patch is never used, so 
> either code there or here must be amended.

Yes, it's just a place holder for future kernel mitigations~
Or I should leave it until we find out these mitigations?

Thanks.

- Jiaxun

>
>>           break;
>>       default:
>>           panic("Unknown Loongson Processor ID!");