Hi Lorenzo, On Tue, May 5, 2020 at 11:29:33, Lorenzo Pieralisi <lorenzo.pieralisi@xxxxxxx> wrote: > On Wed, Apr 01, 2020 at 04:58:13PM -0700, Alan Mikhak wrote: > > From: Alan Mikhak <alan.mikhak@xxxxxxxxxx> > > > > Function dw_pcie_prog_outbound_atu_unroll() does not program the upper > > 32-bit ATU limit register. Since ATU programming functions limit the > > size of the translated region to 4GB by using a u32 size parameter, > > these issues may combine into undefined behavior for resource sizes > > with non-zero upper 32-bits. > > > > For example, a 128GB address space starting at physical CPU address of > > 0x2000000000 with size of 0x2000000000 needs the following values > > programmed into the lower and upper 32-bit limit registers: > > 0x3fffffff in the upper 32-bit limit register > > 0xffffffff in the lower 32-bit limit register > > > > Currently, only the lower 32-bit limit register is programmed with a > > value of 0xffffffff but the upper 32-bit limit register is not being > > programmed. As a result, the upper 32-bit limit register remains at its > > default value after reset of 0x0. > > > > These issues may combine to produce undefined behavior since the ATU > > limit address may be lower than the ATU base address. Programming the > > upper ATU limit address register prevents such undefined behavior despite > > the region size getting truncated due to the 32-bit size limit. > > > > Signed-off-by: Alan Mikhak <alan.mikhak@xxxxxxxxxx> > > --- > > drivers/pci/controller/dwc/pcie-designware.c | 7 +++++-- > > drivers/pci/controller/dwc/pcie-designware.h | 3 ++- > > 2 files changed, 7 insertions(+), 3 deletions(-) > > I would appreciate some feedback and possibly and ACK from DWC > maintainers. Should this go to stable kernels ? It seems so, > let me know if we want to add a stable tag. > > I will merge it, along with: > > https://urldefense.com/v3/__https://patchwork.kernel.org/patch/11468465/__;!!A4F2R9G_pg!NoymSJCWmOx51jB7LdQQAbXFin14nfuVIQNQxROnskLmmGkzFeNOrf8nFWX_-KgsgO87N9M$ > > Lorenzo Sorry for the delay. I just gave the ACK to that patch. For me, it makes sense to me to send it along with the patch that you just referred to the stable kernels. -Gustavo
