On Thu, Apr 22, 2021, Colin King wrote:
> From: Colin Ian King <colin.king@xxxxxxxxxxxxx>
>
> Currently entry->ebx is being zero'd by masking itself with zero.
> Simplify this by just assigning zero, cleans up static analysis
> warning.
>
> Addresses-Coverity: ("Bitwise-and with zero")
> Signed-off-by: Colin Ian King <colin.king@xxxxxxxxxxxxx>
> ---
> arch/x86/kvm/cpuid.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
> index 57744a5d1bc2..9bcc2ff4b232 100644
> --- a/arch/x86/kvm/cpuid.c
> +++ b/arch/x86/kvm/cpuid.c
> @@ -851,7 +851,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
> entry->eax &= SGX_ATTR_DEBUG | SGX_ATTR_MODE64BIT |
> SGX_ATTR_PROVISIONKEY | SGX_ATTR_EINITTOKENKEY |
> SGX_ATTR_KSS;
> - entry->ebx &= 0;
> + entry->ebx = 0;
I 100% understand the code is funky, but using &= is intentional. ebx:eax holds
a 64-bit value that is a effectively a set of feature flags. While the upper
32 bits are extremely unlikely to be used any time soon, if a feature comes
along then the correct behavior would be:
entry->ebx &= SGX_ATTR_FANCY_NEW_FEATURE;
While directly setting entry->ebx would be incorrect. The idea is to set up a
future developer for success so that they don't forget to add the "&".
TL;DR: I'd prefer to keep this as is, even though it's rather ridiculous.
> break;
> /* Intel PT */
> case 0x14:
> --
> 2.30.2
>
