kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Binbin Wu <binbin.wu@linux.intel.com>
To: Sean Christopherson <seanjc@google.com>
Cc: Chao Gao <chao.gao@intel.com>,
	Robert Hoo <robert.hu@linux.intel.com>,
	pbonzini@redhat.com, kvm@vger.kernel.org
Subject: Re: [PATCH v5 2/5] [Trivial]KVM: x86: Explicitly cast ulong to bool in kvm_set_cr3()
Date: Mon, 20 Mar 2023 21:56:15 +0800	[thread overview]
Message-ID: <4ff37ff4-b89e-8683-f6ea-865211ae01d2@linux.intel.com> (raw)
In-Reply-To: <75f628e3-9621-ac9e-a258-33efc7ce56af@linux.intel.com>


On 3/20/2023 8:05 PM, Binbin Wu wrote:
>
> On 3/11/2023 4:22 AM, Sean Christopherson wrote:
>> As Chao pointed out, this does not belong in the LAM series.  And 
>> FWIW, I highly
>> recommend NOT tagging things as Trivial.  If you're wrong and the 
>> patch _isn't_
>> trivial, it only slows things down.  And if you're right, then 
>> expediting the
>> patch can't possibly be necessary.
>>
>> On Fri, Mar 03, 2023, Robert Hoo wrote:
>>> On Thu, 2023-03-02 at 15:24 +0800, Chao Gao wrote:
>>>>> -    bool pcid_enabled = kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE);
>>>>> +    bool pcid_enabled = !!kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE);
>>>>>
>>>>>     if (pcid_enabled) {
>>>>>         skip_tlb_flush = cr3 & X86_CR3_PCID_NOFLUSH;
>>>> pcid_enabled is used only once. You can drop it, i.e.,
>>>>
>>>>     if (kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE)) {
>>>>
>>> Emm, that's actually another point.
>>> Though I won't object so, wouldn't this be compiler optimized?
>>>
>>> And my point was: honor bool type, though in C implemention it's 0 and
>>> !0, it has its own type value: true, false.
>>> Implicit type casting always isn't good habit.
>> I don't disagree, but I also don't particularly want to "fix" one 
>> case while
>> ignoring the many others, e.g. kvm_handle_invpcid() has the exact 
>> same "buggy"
>> pattern.
>>
>> I would be supportive of a patch that adds helpers and then converts 
>> all of the
>> relevant CR0/CR4 checks though...
>
> Hi Sean, I can cook a patch by your suggesion and sent out the patch 
> seperately.

Sean, besides the call of kvm_read_cr0_bits() and kvm_read_cr4_bits(), 
there are also a lot checks in if statement like
if ( cr4 & X86_CR4_XXX )
or
if ( cr0 & X86_CR0_XXX )
I suppose these usages are OK, right?


>
>
>>
>> diff --git a/arch/x86/kvm/kvm_cache_regs.h 
>> b/arch/x86/kvm/kvm_cache_regs.h
>> index 4c91f626c058..6e3cb958afdd 100644
>> --- a/arch/x86/kvm/kvm_cache_regs.h
>> +++ b/arch/x86/kvm/kvm_cache_regs.h
>> @@ -157,6 +157,14 @@ static inline ulong kvm_read_cr0_bits(struct 
>> kvm_vcpu *vcpu, ulong mask)
>>          return vcpu->arch.cr0 & mask;
>>   }
>>   +static __always_inline bool kvm_is_cr0_bit_set(struct kvm_vcpu *vcpu,
>> +                                              unsigned long cr0_bit)
>> +{
>> +       BUILD_BUG_ON(!is_power_of_2(cr0_bit));
>> +
>> +       return !!kvm_read_cr0_bits(vcpu, cr0_bit);
>> +}
>> +
>>   static inline ulong kvm_read_cr0(struct kvm_vcpu *vcpu)
>>   {
>>          return kvm_read_cr0_bits(vcpu, ~0UL);
>> @@ -178,6 +186,14 @@ static inline ulong kvm_read_cr3(struct kvm_vcpu 
>> *vcpu)
>>          return vcpu->arch.cr3;
>>   }
>>   +static __always_inline bool kvm_is_cr4_bit_set(struct kvm_vcpu *vcpu,
>> +                                              unsigned long cr4_bit)
>> +{
>> +       BUILD_BUG_ON(!is_power_of_2(cr4_bit));
>> +
>> +       return !!kvm_read_cr4_bits(vcpu, cr4_bit);
>> +}
>> +

  reply	other threads:[~2023-03-20 13:56 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-27  8:45 [PATCH v5 0/5] Linear Address Masking (LAM) KVM Enabling Robert Hoo
2023-02-27  8:45 ` [PATCH v5 1/5] KVM: x86: Virtualize CR4.LAM_SUP Robert Hoo
2023-03-02  7:17   ` Chao Gao
2023-03-02 12:03     ` Binbin Wu
2023-03-02 13:00     ` Robert Hoo
2023-02-27  8:45 ` [PATCH v5 2/5] [Trivial]KVM: x86: Explicitly cast ulong to bool in kvm_set_cr3() Robert Hoo
2023-03-02  7:24   ` Chao Gao
2023-03-03  3:23     ` Robert Hoo
2023-03-10 20:22       ` Sean Christopherson
2023-03-20 12:05         ` Binbin Wu
2023-03-20 13:56           ` Binbin Wu [this message]
2023-03-21 16:03             ` Sean Christopherson
2023-02-27  8:45 ` [PATCH v5 3/5] KVM: x86: Virtualize CR3.LAM_{U48,U57} Robert Hoo
2023-03-03  6:21   ` Chao Gao
2023-03-03 14:23     ` Robert Hoo
2023-03-03 15:53       ` Chao Gao
2023-03-05  1:31         ` Robert Hoo
2023-03-10 20:12   ` Sean Christopherson
2023-03-20  6:57     ` Binbin Wu
2023-02-27  8:45 ` [PATCH v5 4/5] KVM: x86: emulation: Apply LAM mask when emulating data access in 64-bit mode Robert Hoo
2023-03-02  6:41   ` Binbin Wu
2023-03-02 13:16     ` Robert Hoo
2023-03-03  1:08       ` Binbin Wu
2023-03-03  3:16         ` Robert Hoo
2023-03-03  3:35           ` Binbin Wu
2023-03-03  9:00             ` Robert Hoo
2023-03-03 10:18               ` Binbin Wu
2023-03-10 20:26         ` Sean Christopherson
2023-03-02  8:55   ` Chao Gao
2023-03-02 11:31     ` Binbin Wu
2023-03-10 20:23   ` Sean Christopherson
2023-02-27  8:45 ` [PATCH v5 5/5] KVM: x86: LAM: Expose LAM CPUID to user space VMM Robert Hoo
2023-03-03  6:46   ` Chao Gao

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4ff37ff4-b89e-8683-f6ea-865211ae01d2@linux.intel.com \
    --to=binbin.wu@linux.intel.com \
    --cc=chao.gao@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=robert.hu@linux.intel.com \
    --cc=seanjc@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).