> -----Original Message----- > From: Pratyush Anand [mailto:pratyush.anand@xxxxxxxxx] > Sent: Tuesday, September 29, 2015 4:36 PM > To: Gabriele Paoloni > Cc: Bjorn Helgaas; Jingoo Han; linux-pci@xxxxxxxxxxxxxxx; Wangzhou (B); > Yuanzhichang; Zhudacai; zhangjukuo; qiuzhenfa; Liguozhu (Kenneth) > Subject: Re: [PATCH v5 3/3] PCI: designware: add sanity checks on the > header offset in dw_pcie_cfg_read and dw_pcie_cfg_write > > On Tue, Sep 29, 2015 at 7:27 PM, Gabriele Paoloni > <gabriele.paoloni@xxxxxxxxxx> wrote: > > From: gabriele paoloni <gabriele.paoloni@xxxxxxxxxx> > > > > This patch adds sanity checks on "where" input parameter in > > dw_pcie_cfg_read and dw_pcie_cfg_write. These checks make sure > > that offset passed in by the caller is not in conflict with > > the size of the PCI header field that is being read/written > > > > I am still not convinced that we should doubt the caller..But may be I > am biased in my thoughts... > Since Bjorn has asked about it, so will take it. Yes, I think this should be a maintainer decision as it is a matter of trusting the callers... > > > Signed-off-by: Gabriele Paoloni <gabriele.paoloni@xxxxxxxxxx> > > --- > > drivers/pci/host/pcie-designware.c | 20 ++++++++++++++------ > > 1 file changed, 14 insertions(+), 6 deletions(-) > > > > diff --git a/drivers/pci/host/pcie-designware.c > b/drivers/pci/host/pcie-designware.c > > index d771fa5..719d2cd 100644 > > --- a/drivers/pci/host/pcie-designware.c > > +++ b/drivers/pci/host/pcie-designware.c > > @@ -82,11 +82,15 @@ static inline struct pcie_port > *sys_to_pcie(struct pci_sys_data *sys) > > > > int dw_pcie_cfg_read(void __iomem *addr, int size, u32 *val) > > { > > Wouldn't a single check would have been better > + if ((uintptr_t)addr & (size -1)) > + return PCIBIOS_BAD_REGISTER_NUMBER; Ok agreed will change in v6 Thanks > > ~Pratyush ��.n��������+%������w��{.n�����{���"�)�� jg��� �����ݢj����G���� ���j:+v���w�m������w�������h�����٥
