| Message ID | 1366732717-19657-1-git-send-email-peter.maydell@linaro.org |
|---|---|
| State | Accepted |
| Commit | 536f25e4c77592b936e50728c83894c23f4f61c8 |
| Headers | show |
diff --git a/target-arm/kvm.c b/target-arm/kvm.c index d8acace..b7bdc03 100644 --- a/target-arm/kvm.c +++ b/target-arm/kvm.c @@ -62,8 +62,8 @@ int kvm_arch_init_vcpu(CPUState *cs) r.id = KVM_REG_ARM | KVM_REG_SIZE_U64 | KVM_REG_ARM_VFP | 31; r.addr = (uintptr_t)(&v); ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r); - if (ret == ENOENT) { - return EINVAL; + if (ret == -ENOENT) { + return -EINVAL; } return ret; }
kvm_vcpu_ioctl() returns -ETHING on error, not ETHING -- correct an incorrect check in kvm_arch_init_vcpu(). This would not have had any significant ill-effects -- we would just have propagated the less useful ENOENT up to the caller rather than the more accurate EINVAL in the unlikely case that the kernel didn't have VFP-D32 support. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- Pretty trivial bugfix, noticed while I was doing something to the code in this area... target-arm/kvm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)