| Message ID | 20241025210305.27499-7-terry.bowman@amd.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Bjorn Helgaas |
| Headers | show |
| Series | Enable CXL PCIe port protocol error handling and logging | expand |
On Fri, 25 Oct 2024 16:02:57 -0500 Terry Bowman <terry.bowman@amd.com> wrote: > The AER service driver's aer_get_device_error_info() function doesn't read > uncorrectable (UCE) fatal error status from PCIe upstream port devices, > including CXL upstream switch ports. As a result, fatal errors are not > logged or handled as needed for CXL PCIe upstream switch port devices. > > Update the aer_get_device_error_info() function to read the UCE fatal > status for all CXL PCIe port devices. > > The fatal error status will be used in future patches implementing > CXL PCIe port uncorrectable error handling and logging. > > Signed-off-by: Terry Bowman <terry.bowman@amd.com> I assume this was previously not done because the upstream port requires a healthy link and maybe the error indicates we don't have one. So I'd imagine this change may have a bad effect on PCIe devices even if we know it's fine CXL ones in the case of certain protocol errors. Also, does the error log stuff that follows make much sense for an upstream port? > --- > drivers/pci/pcie/aer.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c > index 1d3e5b929661..d772f123c6a2 100644 > --- a/drivers/pci/pcie/aer.c > +++ b/drivers/pci/pcie/aer.c > @@ -1250,6 +1250,7 @@ int aer_get_device_error_info(struct pci_dev *dev, struct aer_err_info *info) > } else if (type == PCI_EXP_TYPE_ROOT_PORT || > type == PCI_EXP_TYPE_RC_EC || > type == PCI_EXP_TYPE_DOWNSTREAM || > + type == PCI_EXP_TYPE_UPSTREAM || > info->severity == AER_NONFATAL) { > > /* Link is still healthy for IO reads */
On 10/25/24 2:02 PM, Terry Bowman wrote: > The AER service driver's aer_get_device_error_info() function doesn't read > uncorrectable (UCE) fatal error status from PCIe upstream port devices, > including CXL upstream switch ports. As a result, fatal errors are not > logged or handled as needed for CXL PCIe upstream switch port devices. > > Update the aer_get_device_error_info() function to read the UCE fatal > status for all CXL PCIe port devices. > > The fatal error status will be used in future patches implementing > CXL PCIe port uncorrectable error handling and logging. > > Signed-off-by: Terry Bowman <terry.bowman@amd.com> > --- > drivers/pci/pcie/aer.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c > index 1d3e5b929661..d772f123c6a2 100644 > --- a/drivers/pci/pcie/aer.c > +++ b/drivers/pci/pcie/aer.c > @@ -1250,6 +1250,7 @@ int aer_get_device_error_info(struct pci_dev *dev, struct aer_err_info *info) > } else if (type == PCI_EXP_TYPE_ROOT_PORT || > type == PCI_EXP_TYPE_RC_EC || > type == PCI_EXP_TYPE_DOWNSTREAM || > + type == PCI_EXP_TYPE_UPSTREAM || At minimal we probably should do something like (pcie_is_cxl(dev) && type == PCI_EXP_TYPE_UPSTREAM) instead so we don't regress the original PCI behavior? > info->severity == AER_NONFATAL) { > > /* Link is still healthy for IO reads */
Hi Dave, On 10/31/2024 11:58 AM, Dave Jiang wrote: > > On 10/25/24 2:02 PM, Terry Bowman wrote: >> The AER service driver's aer_get_device_error_info() function doesn't read >> uncorrectable (UCE) fatal error status from PCIe upstream port devices, >> including CXL upstream switch ports. As a result, fatal errors are not >> logged or handled as needed for CXL PCIe upstream switch port devices. >> >> Update the aer_get_device_error_info() function to read the UCE fatal >> status for all CXL PCIe port devices. >> >> The fatal error status will be used in future patches implementing >> CXL PCIe port uncorrectable error handling and logging. >> >> Signed-off-by: Terry Bowman <terry.bowman@amd.com> >> --- >> drivers/pci/pcie/aer.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c >> index 1d3e5b929661..d772f123c6a2 100644 >> --- a/drivers/pci/pcie/aer.c >> +++ b/drivers/pci/pcie/aer.c >> @@ -1250,6 +1250,7 @@ int aer_get_device_error_info(struct pci_dev *dev, struct aer_err_info *info) >> } else if (type == PCI_EXP_TYPE_ROOT_PORT || >> type == PCI_EXP_TYPE_RC_EC || >> type == PCI_EXP_TYPE_DOWNSTREAM || >> + type == PCI_EXP_TYPE_UPSTREAM || > At minimal we probably should do something like > (pcie_is_cxl(dev) && type == PCI_EXP_TYPE_UPSTREAM) > instead so we don't regress the original PCI behavior? Good Idea. I'll change the condition to what you recommend. Regards, Terry
diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c index 1d3e5b929661..d772f123c6a2 100644 --- a/drivers/pci/pcie/aer.c +++ b/drivers/pci/pcie/aer.c @@ -1250,6 +1250,7 @@ int aer_get_device_error_info(struct pci_dev *dev, struct aer_err_info *info) } else if (type == PCI_EXP_TYPE_ROOT_PORT || type == PCI_EXP_TYPE_RC_EC || type == PCI_EXP_TYPE_DOWNSTREAM || + type == PCI_EXP_TYPE_UPSTREAM || info->severity == AER_NONFATAL) { /* Link is still healthy for IO reads */
The AER service driver's aer_get_device_error_info() function doesn't read uncorrectable (UCE) fatal error status from PCIe upstream port devices, including CXL upstream switch ports. As a result, fatal errors are not logged or handled as needed for CXL PCIe upstream switch port devices. Update the aer_get_device_error_info() function to read the UCE fatal status for all CXL PCIe port devices. The fatal error status will be used in future patches implementing CXL PCIe port uncorrectable error handling and logging. Signed-off-by: Terry Bowman <terry.bowman@amd.com> --- drivers/pci/pcie/aer.c | 1 + 1 file changed, 1 insertion(+)