On Tue, Jan 03, 2017 at 09:09:49AM +0100, Jan Rüth wrote: > This patch fixes a null pointer dereference during PCI bus enumeration > when ASPM is off. On an IBM x3850 8664 this bug causes the kernel to > halt, this behavior did not appear in 3.10, so this is a regression. > > pcie_aspm_sanity_check should only be called if ASPM is on. > > Signed-off-by: Jan Rueth <rueth@xxxxxxxxxxxxxxxxxxxxx> We have https://bugzilla.kernel.org/show_bug.cgi?id=187731 for this issue. Per the dmesg attached there, you're booting with "pcie_aspm=off". This patch fixes the NULL pointer dereference in that case, but I'm concerned that we may still trip over it if you boot without "pcie_aspm=off". I want to make sure we fix both cases. Can you please try a boot with this patch but without "pcie_aspm=off"? > --- > drivers/pci/pcie/aspm.c | +++-- > 1 file changed, 3 insertions(+), 2 deletions(-) > > diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c > index f981129..e758b56 100644 > --- a/drivers/pci/pcie/aspm.c > +++ b/drivers/pci/pcie/aspm.c > @@ -552,11 +552,12 @@ static struct pcie_link_state > *alloc_pcie_link_state(struct pci_dev *pdev) > void pcie_aspm_init_link_state(struct pci_dev *pdev) > { > struct pcie_link_state *link; > - int blacklist = !!pcie_aspm_sanity_check(pdev); > - > + int blacklist; > if (!aspm_support_enabled) > return; > > + blacklist = !!pcie_aspm_sanity_check(pdev); > + > if (pdev->link_state) > return; > -- > To unsubscribe from this list: send the line "unsubscribe linux-pci" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html
