]> Git Repo - J-linux.git/commitdiff
x86/vmware: Use BIT() macro for shifting
authorShreenidhi Shedi <[email protected]>
Wed, 1 Jun 2022 10:18:20 +0000 (15:48 +0530)
committerBorislav Petkov <[email protected]>
Wed, 22 Jun 2022 09:23:14 +0000 (11:23 +0200)
VMWARE_CMD_VCPU_RESERVED is bit 31 and that would mean undefined
behavior when shifting an int but the kernel is built with
-fno-strict-overflow which will wrap around using two's complement.

Use the BIT() macro to improve readability and avoid any potential
overflow confusion because it uses an unsigned long.

  [ bp: Clarify commit message. ]

Signed-off-by: Shreenidhi Shedi <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Srivatsa S. Bhat (VMware) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
arch/x86/kernel/cpu/vmware.c

index c04b933f48d38257e1ca241875a10581b1c92121..02039ec3597d72474b797a5ef47c5abe6e1a9065 100644 (file)
@@ -476,8 +476,8 @@ static bool __init vmware_legacy_x2apic_available(void)
 {
        uint32_t eax, ebx, ecx, edx;
        VMWARE_CMD(GETVCPU_INFO, eax, ebx, ecx, edx);
-       return (eax & (1 << VMWARE_CMD_VCPU_RESERVED)) == 0 &&
-              (eax & (1 << VMWARE_CMD_LEGACY_X2APIC)) != 0;
+       return !(eax & BIT(VMWARE_CMD_VCPU_RESERVED)) &&
+               (eax & BIT(VMWARE_CMD_LEGACY_X2APIC));
 }
 
 #ifdef CONFIG_AMD_MEM_ENCRYPT
This page took 0.050153 seconds and 4 git commands to generate.