linux-kernel - Re: [PATCH v5 2/9] x86/split_lock: Avoid runtime reads of the TEST

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]

Date:   Tue, 24 Mar 2020 09:16:38 +0800
From:   Xiaoyao Li <xiaoyao.li@...el.com>
To:     Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        hpa@...or.com, Paolo Bonzini <pbonzini@...hat.com>,
        Sean Christopherson <sean.j.christopherson@...el.com>,
        kvm@...r.kernel.org, x86@...nel.org, linux-kernel@...r.kernel.org
Cc:     Andy Lutomirski <luto@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Arvind Sankar <nivedita@...m.mit.edu>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Tony Luck <tony.luck@...el.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Jim Mattson <jmattson@...gle.com>
Subject: Re: [PATCH v5 2/9] x86/split_lock: Avoid runtime reads of the
 TEST_CTRL MSR

On 3/24/2020 1:06 AM, Thomas Gleixner wrote:
> Xiaoyao Li <xiaoyao.li@...el.com> writes:
>> +/*
>> + * Soft copy of MSR_TEST_CTRL initialized when we first read the
>> + * MSR. Used at runtime to avoid using rdmsr again just to collect
>> + * the reserved bits in the MSR. We assume reserved bits are the
>> + * same on all CPUs.
>> + */
>> +static u64 test_ctrl_val;
>> +
>>   /*
>>    * Locking is not required at the moment because only bit 29 of this
>>    * MSR is implemented and locking would not prevent that the operation
>> @@ -1027,16 +1035,14 @@ static void __init split_lock_setup(void)
>>    */
>>   static void __sld_msr_set(bool on)
>>   {
>> -	u64 test_ctrl_val;
>> -
>> -	rdmsrl(MSR_TEST_CTRL, test_ctrl_val);
>> +	u64 val = test_ctrl_val;
>>   
>>   	if (on)
>> -		test_ctrl_val |= MSR_TEST_CTRL_SPLIT_LOCK_DETECT;
>> +		val |= MSR_TEST_CTRL_SPLIT_LOCK_DETECT;
>>   	else
>> -		test_ctrl_val &= ~MSR_TEST_CTRL_SPLIT_LOCK_DETECT;
>> +		val &= ~MSR_TEST_CTRL_SPLIT_LOCK_DETECT;
>>   
>> -	wrmsrl(MSR_TEST_CTRL, test_ctrl_val);
>> +	wrmsrl(MSR_TEST_CTRL, val);
>>   }
>>   
>>   /*
>> @@ -1048,11 +1054,13 @@ static void __sld_msr_set(bool on)
>>    */
>>   static void split_lock_init(struct cpuinfo_x86 *c)
>>   {
>> -	u64 test_ctrl_val;
>> +	u64 val;
>>   
>> -	if (rdmsrl_safe(MSR_TEST_CTRL, &test_ctrl_val))
>> +	if (rdmsrl_safe(MSR_TEST_CTRL, &val))
>>   		goto msr_broken;
>>   
>> +	test_ctrl_val = val;
>> +
>>   	switch (sld_state) {
>>   	case sld_off:
>>   		if (wrmsrl_safe(MSR_TEST_CTRL, test_ctrl_val & ~MSR_TEST_CTRL_SPLIT_LOCK_DETECT))
> 
> That's just broken. Simply because
> 
>         case sld_warn:
>         case sld_fatal:
> 
> set the split lock detect bit, but the cache variable has it cleared
> unless it was set at boot time already.

The test_ctrl_val is not to cache the value of MSR_TEST_CTRL, but cache 
the reserved/unused bits other than MSR_TEST_CTRL_SPLIT_LOCK_DETECT bit.

> Thanks,
> 
>          tglx
>