People have been running W=1 to catch additional compilation warnings with an
expectation that not all of warnings will be fixed, but the linux-next commit
ead68df94d24 ("KVM: x86: enable -Werror") breaks the build for them.
arch/x86/kvm/../../../virt/kvm/kvm_main.c:663:12: error: no previous prototype
for ‘kvm_arch_post_init_vm’ [-Werror=missing-prototypes]
int __weak kvm_arch_post_init_vm(struct kvm *kvm)
^~~~~~~~~~~~~~~~~~~~~
arch/x86/kvm/../../../virt/kvm/kvm_main.c:672:13: error: no previous prototype
for ‘kvm_arch_pre_destroy_vm’ [-Werror=missing-prototypes]
void __weak kvm_arch_pre_destroy_vm(struct kvm *kvm)
On 13/02/20 14:50, Qian Cai wrote:
> People have been running W=1 to catch additional compilation warnings with an
> expectation that not all of warnings will be fixed, but the linux-next commit
> ead68df94d24 ("KVM: x86: enable -Werror") breaks the build for them.
>
> arch/x86/kvm/../../../virt/kvm/kvm_main.c:663:12: error: no previous prototype
> for ‘kvm_arch_post_init_vm’ [-Werror=missing-prototypes]
> int __weak kvm_arch_post_init_vm(struct kvm *kvm)
> ^~~~~~~~~~~~~~~~~~~~~
> arch/x86/kvm/../../../virt/kvm/kvm_main.c:672:13: error: no previous prototype
> for ‘kvm_arch_pre_destroy_vm’ [-Werror=missing-prototypes]
> void __weak kvm_arch_pre_destroy_vm(struct kvm *kvm)
Thanks, we already found two bugs, one of them real (it's from
-Wtype-limits). But I'll leave out this commit for now from the pull
request, of course.
Paolo