On Mon, 2015-12-28 at 00:16 -0500, Joshua Otto wrote: Please check MAINTAINERS (perhaps using ./scripts/get_maintainers.pl) to determine the maintainers in order to CC them.
(Hrm, I see now in your cover letter you have, I wonder why Jan et al are missing?) > From: Harley Armstrong <hjarm...@uwaterloo.ca> > > Fix CID 1343302 by widening a constant used with uart_param.reg_shift to > avoid overflow for large values of reg_shift. A reg_shift large enough to actually expose this would be infeasibly large (since it would imply a UART taking practically the entire virtual address space of the processor). So while Coverity is likely correct here, it is probably also a bit misguided in the context. I don't especially object to this change as means to quieten coverity, but perhaps checking for some sane limit to reg_shift would also serve to quieten coverity? That would also avoid the need to check for overflow on the multiplication, assuming a suitable sane limit was chosen. > Signed-off-by: Harley Armstrong <hjarm...@uwaterloo.ca> > --- > xen/drivers/char/ns16550.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/xen/drivers/char/ns16550.c b/xen/drivers/char/ns16550.c > index bc24015..546bba1 100644 > --- a/xen/drivers/char/ns16550.c > +++ b/xen/drivers/char/ns16550.c > @@ -913,7 +913,7 @@ pci_uart_config(struct ns16550 *uart, bool_t > skip_amt, unsigned int bar_idx) > * Force length of mmio region to be at least > * 8 bytes times (1 << reg_shift) > */ > - if ( size < (0x8 * (1 << > uart_param[p].reg_shift)) ) > + if ( size < (0x8 * (1ull << > uart_param[p].reg_shift)) ) It looks from the surrounding code like ... < (0x8 * ((u64)1 << uart_param[p].reg_shift)) ) would be the preferred way of tackling this. > continue; > > if ( bar_idx >= uart_param[p].max_bars ) _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
