Messages in this thread | | | Date | Thu, 15 Nov 2012 16:03:31 -0800 | From | Greg KH <> | Subject | Re: [PATCH 11/12] VMCI: host side driver implementation. |
| |
On Wed, Nov 07, 2012 at 10:42:53AM -0800, George Zhang wrote: > +/* > + * Sets up a given context for notify to work. Calls drv_map_bool_ptr() > + * which maps the notify boolean in user VA in kernel space. > + */ > +static int vmci_host_setup_notify(struct vmci_ctx *context, > + unsigned long uva) > +{ > + struct page *page; > + int retval; > + > + if (context->notify_page) { > + pr_devel("%s: Notify mechanism is already set up.\n", __func__); > + return VMCI_ERROR_DUPLICATE_ENTRY; > + } > + > + if (!access_ok(VERIFY_WRITE, (void __user *)uva, sizeof(bool))) > + return VMCI_ERROR_GENERIC;
This line causes sparse to complain. The odds that userspace knows what gcc is using for "bool" is pretty low.
Please fix.
thanks,
greg k-h
| |