| Message ID | 20220127025418.1989642-1-kai.heng.feng@canonical.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Bjorn Helgaas |
| Headers | show |
| Series | [v2,1/2] PCI/AER: Disable AER service when link is in L2/L3 ready, L2 and L3 state | expand |
On Thu, Jan 27, 2022 at 10:54:17AM +0800, Kai-Heng Feng wrote: > Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in > hint") enables ACS, and some platforms lose its NVMe after resume from > S3: > [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 > [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected > [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) > [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 > [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) > [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message > [ 50.947843] nvme nvme0: frozen state error detected, reset controller > > It happens right after ACS gets enabled during resume. > > There's another case, when Thunderbolt reaches D3cold: > [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 > [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) > [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 > [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) > [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 > [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) > [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) > [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed > > So disable AER service to avoid the noises from turning power rails > on/off when the device is in low power states (D3hot and D3cold), as > PCIe spec "5.2 Link State Power Management" states that TLP and DLLP > transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold > with aux power) and L3 (D3cold). > > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=209149 > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=215453 > Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") > Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com> Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
On 2022/1/27 10:54, Kai-Heng Feng wrote: > Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in > hint") enables ACS, and some platforms lose its NVMe after resume from > S3: > [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 > [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected > [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) > [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 > [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) > [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message > [ 50.947843] nvme nvme0: frozen state error detected, reset controller > > It happens right after ACS gets enabled during resume. > > There's another case, when Thunderbolt reaches D3cold: > [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 > [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) > [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 > [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) > [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 > [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) > [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) > [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed > > So disable AER service to avoid the noises from turning power rails > on/off when the device is in low power states (D3hot and D3cold), as > PCIe spec "5.2 Link State Power Management" states that TLP and DLLP > transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold > with aux power) and L3 (D3cold). > > Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=209149 > Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=215453 > Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") I don't know what this fix has to do with the commit 50310600ebda. Commit 50310600ebda only makes sure that PCI ACS is enabled whenever Intel IOMMU is on. Before this commit, PCI ACS could also be enabled and result in the same problem. Or anything I missed? Best regards, baolu
On Thu, Jan 27, 2022 at 3:01 PM Lu Baolu <baolu.lu@linux.intel.com> wrote: > > On 2022/1/27 10:54, Kai-Heng Feng wrote: > > Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in > > hint") enables ACS, and some platforms lose its NVMe after resume from > > S3: > > [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 > > [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected > > [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) > > [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 > > [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) > > [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message > > [ 50.947843] nvme nvme0: frozen state error detected, reset controller > > > > It happens right after ACS gets enabled during resume. > > > > There's another case, when Thunderbolt reaches D3cold: > > [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 > > [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) > > [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 > > [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) > > [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 > > [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) > > [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) > > [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed > > > > So disable AER service to avoid the noises from turning power rails > > on/off when the device is in low power states (D3hot and D3cold), as > > PCIe spec "5.2 Link State Power Management" states that TLP and DLLP > > transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold > > with aux power) and L3 (D3cold). > > > > Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=209149 > > Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=215453 > > Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") > > I don't know what this fix has to do with the commit 50310600ebda. Commit 50310600ebda only exposed the underlying issue. Do you think "Fixes:" tag should change to other commits? > Commit 50310600ebda only makes sure that PCI ACS is enabled whenever > Intel IOMMU is on. Before this commit, PCI ACS could also be enabled > and result in the same problem. Or anything I missed? The system in question didn't enable ACS before commit 50310600ebda. Kai-Heng > > Best regards, > baolu
On 1/27/22 7:14 PM, Kai-Heng Feng wrote: > On Thu, Jan 27, 2022 at 3:01 PM Lu Baolu <baolu.lu@linux.intel.com> wrote: >> >> On 2022/1/27 10:54, Kai-Heng Feng wrote: >>> Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in >>> hint") enables ACS, and some platforms lose its NVMe after resume from >>> S3: >>> [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 >>> [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected >>> [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) >>> [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 >>> [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) >>> [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message >>> [ 50.947843] nvme nvme0: frozen state error detected, reset controller >>> >>> It happens right after ACS gets enabled during resume. >>> >>> There's another case, when Thunderbolt reaches D3cold: >>> [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 >>> [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) >>> [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 >>> [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) >>> [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 >>> [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) >>> [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) >>> [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed >>> >>> So disable AER service to avoid the noises from turning power rails >>> on/off when the device is in low power states (D3hot and D3cold), as >>> PCIe spec "5.2 Link State Power Management" states that TLP and DLLP >>> transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold >>> with aux power) and L3 (D3cold). >>> >>> Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=209149 >>> Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=215453 >>> Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") >> >> I don't know what this fix has to do with the commit 50310600ebda. > > Commit 50310600ebda only exposed the underlying issue. Do you think > "Fixes:" tag should change to other commits? > >> Commit 50310600ebda only makes sure that PCI ACS is enabled whenever >> Intel IOMMU is on. Before this commit, PCI ACS could also be enabled >> and result in the same problem. Or anything I missed? > > The system in question didn't enable ACS before commit 50310600ebda. This commit exposed the issue on your configuration doesn't mean the fix should be back ported as far as that commit. I believe if you add intel-iommu=on in the kernel parameter, the issue still exists even you revert commit 50310600ebda or checkout a tag before it. Best regards, baolu
On Fri, Jan 28, 2022 at 10:57 AM Lu Baolu <baolu.lu@linux.intel.com> wrote: > > On 1/27/22 7:14 PM, Kai-Heng Feng wrote: > > On Thu, Jan 27, 2022 at 3:01 PM Lu Baolu <baolu.lu@linux.intel.com> wrote: > >> > >> On 2022/1/27 10:54, Kai-Heng Feng wrote: > >>> Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in > >>> hint") enables ACS, and some platforms lose its NVMe after resume from > >>> S3: > >>> [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 > >>> [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected > >>> [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) > >>> [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 > >>> [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) > >>> [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message > >>> [ 50.947843] nvme nvme0: frozen state error detected, reset controller > >>> > >>> It happens right after ACS gets enabled during resume. > >>> > >>> There's another case, when Thunderbolt reaches D3cold: > >>> [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 > >>> [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) > >>> [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 > >>> [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) > >>> [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 > >>> [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) > >>> [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) > >>> [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed > >>> > >>> So disable AER service to avoid the noises from turning power rails > >>> on/off when the device is in low power states (D3hot and D3cold), as > >>> PCIe spec "5.2 Link State Power Management" states that TLP and DLLP > >>> transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold > >>> with aux power) and L3 (D3cold). > >>> > >>> Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=209149 > >>> Bugzilla:https://bugzilla.kernel.org/show_bug.cgi?id=215453 > >>> Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") > >> > >> I don't know what this fix has to do with the commit 50310600ebda. > > > > Commit 50310600ebda only exposed the underlying issue. Do you think > > "Fixes:" tag should change to other commits? > > > >> Commit 50310600ebda only makes sure that PCI ACS is enabled whenever > >> Intel IOMMU is on. Before this commit, PCI ACS could also be enabled > >> and result in the same problem. Or anything I missed? > > > > The system in question didn't enable ACS before commit 50310600ebda. > > This commit exposed the issue on your configuration doesn't mean the > fix should be back ported as far as that commit. I believe if you add > intel-iommu=on in the kernel parameter, the issue still exists even you > revert commit 50310600ebda or checkout a tag before it. That's true. I guess it's better to drop the "Fixes:" tag. Bjorn, should I send another version of it? Kai-Heng > > Best regards, > baolu
On 1/26/22 6:54 PM, Kai-Heng Feng wrote: > Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in > hint") enables ACS, and some platforms lose its NVMe after resume from Why enabling ACS makes platform lose NVMe? Can you add more details about the problem? > S3: > [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 > [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected > [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) > [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 > [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) > [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message > [ 50.947843] nvme nvme0: frozen state error detected, reset controller > > It happens right after ACS gets enabled during resume. > > There's another case, when Thunderbolt reaches D3cold: > [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 > [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) > [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 > [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) > [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 > [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) no callback message means one or more devices in the given port does not support error handler. How is this related to ACS? > [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) > [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed > > So disable AER service to avoid the noises from turning power rails > on/off when the device is in low power states (D3hot and D3cold), as > PCIe spec "5.2 Link State Power Management" states that TLP and DLLP > transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold > with aux power) and L3 (D3cold). > > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=209149 > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=215453 > Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") > Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com> > --- > v2: > - Wording change. > > drivers/pci/pcie/aer.c | 31 +++++++++++++++++++++++++------ > 1 file changed, 25 insertions(+), 6 deletions(-) > > diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c > index 9fa1f97e5b270..e4e9d4a3098d7 100644 > --- a/drivers/pci/pcie/aer.c > +++ b/drivers/pci/pcie/aer.c > @@ -1367,6 +1367,22 @@ static int aer_probe(struct pcie_device *dev) > return 0; > } > > +static int aer_suspend(struct pcie_device *dev) > +{ > + struct aer_rpc *rpc = get_service_data(dev); > + > + aer_disable_rootport(rpc); > + return 0; > +} > + > +static int aer_resume(struct pcie_device *dev) > +{ > + struct aer_rpc *rpc = get_service_data(dev); > + > + aer_enable_rootport(rpc); > + return 0; > +} > + > /** > * aer_root_reset - reset Root Port hierarchy, RCEC, or RCiEP > * @dev: pointer to Root Port, RCEC, or RCiEP > @@ -1433,12 +1449,15 @@ static pci_ers_result_t aer_root_reset(struct pci_dev *dev) > } > > static struct pcie_port_service_driver aerdriver = { > - .name = "aer", > - .port_type = PCIE_ANY_PORT, > - .service = PCIE_PORT_SERVICE_AER, > - > - .probe = aer_probe, > - .remove = aer_remove, > + .name = "aer", > + .port_type = PCIE_ANY_PORT, > + .service = PCIE_PORT_SERVICE_AER, > + .probe = aer_probe, > + .suspend = aer_suspend, > + .resume = aer_resume, > + .runtime_suspend = aer_suspend, > + .runtime_resume = aer_resume, > + .remove = aer_remove, > }; > > /**
On Sun, Mar 20, 2022 at 4:38 AM Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@linux.intel.com> wrote: > > > > On 1/26/22 6:54 PM, Kai-Heng Feng wrote: > > Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in > > hint") enables ACS, and some platforms lose its NVMe after resume from > > Why enabling ACS makes platform lose NVMe? Can you add more details > about the problem? I don't have a hardware analyzer, so the only detail I can provide is the symptom. I believe the affected system was sent Intel, and there wasn't any feedback since then. > > > S3: > > [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 > > [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected > > [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) > > [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 > > [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) > > [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message > > [ 50.947843] nvme nvme0: frozen state error detected, reset controller > > > > It happens right after ACS gets enabled during resume. > > > > There's another case, when Thunderbolt reaches D3cold: > > [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 > > [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) > > [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 > > [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) > > [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 > > [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) > > no callback message means one or more devices in the given port does not > support error handler. How is this related to ACS? This case is about D3cold, not related to ACS. And no error_detected is just part of the message. The whole AER message is more important. Kai-Heng > > > [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) > > [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed > > > > So disable AER service to avoid the noises from turning power rails > > on/off when the device is in low power states (D3hot and D3cold), as > > PCIe spec "5.2 Link State Power Management" states that TLP and DLLP > > transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold > > with aux power) and L3 (D3cold). > > > > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=209149 > > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=215453 > > Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") > > Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com> > > --- > > v2: > > - Wording change. > > > > drivers/pci/pcie/aer.c | 31 +++++++++++++++++++++++++------ > > 1 file changed, 25 insertions(+), 6 deletions(-) > > > > diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c > > index 9fa1f97e5b270..e4e9d4a3098d7 100644 > > --- a/drivers/pci/pcie/aer.c > > +++ b/drivers/pci/pcie/aer.c > > @@ -1367,6 +1367,22 @@ static int aer_probe(struct pcie_device *dev) > > return 0; > > } > > > > +static int aer_suspend(struct pcie_device *dev) > > +{ > > + struct aer_rpc *rpc = get_service_data(dev); > > + > > + aer_disable_rootport(rpc); > > + return 0; > > +} > > + > > +static int aer_resume(struct pcie_device *dev) > > +{ > > + struct aer_rpc *rpc = get_service_data(dev); > > + > > + aer_enable_rootport(rpc); > > + return 0; > > +} > > + > > /** > > * aer_root_reset - reset Root Port hierarchy, RCEC, or RCiEP > > * @dev: pointer to Root Port, RCEC, or RCiEP > > @@ -1433,12 +1449,15 @@ static pci_ers_result_t aer_root_reset(struct pci_dev *dev) > > } > > > > static struct pcie_port_service_driver aerdriver = { > > - .name = "aer", > > - .port_type = PCIE_ANY_PORT, > > - .service = PCIE_PORT_SERVICE_AER, > > - > > - .probe = aer_probe, > > - .remove = aer_remove, > > + .name = "aer", > > + .port_type = PCIE_ANY_PORT, > > + .service = PCIE_PORT_SERVICE_AER, > > + .probe = aer_probe, > > + .suspend = aer_suspend, > > + .resume = aer_resume, > > + .runtime_suspend = aer_suspend, > > + .runtime_resume = aer_resume, > > + .remove = aer_remove, > > }; > > > > /** > > -- > Sathyanarayanan Kuppuswamy > Linux Kernel Developer
On 3/20/22 7:38 PM, Kai-Heng Feng wrote: > On Sun, Mar 20, 2022 at 4:38 AM Sathyanarayanan Kuppuswamy > <sathyanarayanan.kuppuswamy@linux.intel.com> wrote: >> >> >> >> On 1/26/22 6:54 PM, Kai-Heng Feng wrote: >>> Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in >>> hint") enables ACS, and some platforms lose its NVMe after resume from >> >> Why enabling ACS makes platform lose NVMe? Can you add more details >> about the problem? > > I don't have a hardware analyzer, so the only detail I can provide is > the symptom. > I believe the affected system was sent Intel, and there wasn't any > feedback since then. Since your commit log refers to ACS, I think first we need to understand following points. 1. Why we get ACSViol during S3 resume. Is this just a noise? 2. Why AER recovery fails? 3. Is this common for all platforms, or only happens in your test platform? If you are not clear about above points, I think you can submit this patch as adding suspend/resume support to AER/DPC driver and not include the issue about ACS. From your commit log, the problem is not very clear. > >> >>> S3: >>> [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 >>> [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected >>> [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) >>> [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 >>> [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) >>> [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message >>> [ 50.947843] nvme nvme0: frozen state error detected, reset controller >>> >>> It happens right after ACS gets enabled during resume. >>> >>> There's another case, when Thunderbolt reaches D3cold: >>> [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 >>> [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) >>> [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 >>> [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) >>> [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 >>> [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) >> >> no callback message means one or more devices in the given port does not >> support error handler. How is this related to ACS? > > This case is about D3cold, not related to ACS. > And no error_detected is just part of the message. The whole AER > message is more important. > > Kai-Heng > >> >>> [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) >>> [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed >>> >>> So disable AER service to avoid the noises from turning power rails >>> on/off when the device is in low power states (D3hot and D3cold), as >>> PCIe spec "5.2 Link State Power Management" states that TLP and DLLP >>> transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold >>> with aux power) and L3 (D3cold). >>> >>> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=209149 >>> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=215453 >>> Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") >>> Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com> >>> --- >>> v2: >>> - Wording change. >>> >>> drivers/pci/pcie/aer.c | 31 +++++++++++++++++++++++++------ >>> 1 file changed, 25 insertions(+), 6 deletions(-) >>> >>> diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c >>> index 9fa1f97e5b270..e4e9d4a3098d7 100644 >>> --- a/drivers/pci/pcie/aer.c >>> +++ b/drivers/pci/pcie/aer.c >>> @@ -1367,6 +1367,22 @@ static int aer_probe(struct pcie_device *dev) >>> return 0; >>> } >>> >>> +static int aer_suspend(struct pcie_device *dev) >>> +{ >>> + struct aer_rpc *rpc = get_service_data(dev); >>> + >>> + aer_disable_rootport(rpc); >>> + return 0; >>> +} >>> + >>> +static int aer_resume(struct pcie_device *dev) >>> +{ >>> + struct aer_rpc *rpc = get_service_data(dev); >>> + >>> + aer_enable_rootport(rpc); >>> + return 0; >>> +} >>> + >>> /** >>> * aer_root_reset - reset Root Port hierarchy, RCEC, or RCiEP >>> * @dev: pointer to Root Port, RCEC, or RCiEP >>> @@ -1433,12 +1449,15 @@ static pci_ers_result_t aer_root_reset(struct pci_dev *dev) >>> } >>> >>> static struct pcie_port_service_driver aerdriver = { >>> - .name = "aer", >>> - .port_type = PCIE_ANY_PORT, >>> - .service = PCIE_PORT_SERVICE_AER, >>> - >>> - .probe = aer_probe, >>> - .remove = aer_remove, >>> + .name = "aer", >>> + .port_type = PCIE_ANY_PORT, >>> + .service = PCIE_PORT_SERVICE_AER, >>> + .probe = aer_probe, >>> + .suspend = aer_suspend, >>> + .resume = aer_resume, >>> + .runtime_suspend = aer_suspend, >>> + .runtime_resume = aer_resume, >>> + .remove = aer_remove, >>> }; >>> >>> /** >> >> -- >> Sathyanarayanan Kuppuswamy >> Linux Kernel Developer
diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c index 9fa1f97e5b270..e4e9d4a3098d7 100644 --- a/drivers/pci/pcie/aer.c +++ b/drivers/pci/pcie/aer.c @@ -1367,6 +1367,22 @@ static int aer_probe(struct pcie_device *dev) return 0; } +static int aer_suspend(struct pcie_device *dev) +{ + struct aer_rpc *rpc = get_service_data(dev); + + aer_disable_rootport(rpc); + return 0; +} + +static int aer_resume(struct pcie_device *dev) +{ + struct aer_rpc *rpc = get_service_data(dev); + + aer_enable_rootport(rpc); + return 0; +} + /** * aer_root_reset - reset Root Port hierarchy, RCEC, or RCiEP * @dev: pointer to Root Port, RCEC, or RCiEP @@ -1433,12 +1449,15 @@ static pci_ers_result_t aer_root_reset(struct pci_dev *dev) } static struct pcie_port_service_driver aerdriver = { - .name = "aer", - .port_type = PCIE_ANY_PORT, - .service = PCIE_PORT_SERVICE_AER, - - .probe = aer_probe, - .remove = aer_remove, + .name = "aer", + .port_type = PCIE_ANY_PORT, + .service = PCIE_PORT_SERVICE_AER, + .probe = aer_probe, + .suspend = aer_suspend, + .resume = aer_resume, + .runtime_suspend = aer_suspend, + .runtime_resume = aer_resume, + .remove = aer_remove, }; /**
Commit 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") enables ACS, and some platforms lose its NVMe after resume from S3: [ 50.947816] pcieport 0000:00:1b.0: DPC: containment event, status:0x1f01 source:0x0000 [ 50.947817] pcieport 0000:00:1b.0: DPC: unmasked uncorrectable error detected [ 50.947829] pcieport 0000:00:1b.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Receiver ID) [ 50.947830] pcieport 0000:00:1b.0: device [8086:06ac] error status/mask=00200000/00010000 [ 50.947831] pcieport 0000:00:1b.0: [21] ACSViol (First) [ 50.947841] pcieport 0000:00:1b.0: AER: broadcast error_detected message [ 50.947843] nvme nvme0: frozen state error detected, reset controller It happens right after ACS gets enabled during resume. There's another case, when Thunderbolt reaches D3cold: [ 30.100211] pcieport 0000:00:1d.0: AER: Uncorrected (Non-Fatal) error received: 0000:00:1d.0 [ 30.100251] pcieport 0000:00:1d.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) [ 30.100256] pcieport 0000:00:1d.0: device [8086:7ab0] error status/mask=00100000/00004000 [ 30.100262] pcieport 0000:00:1d.0: [20] UnsupReq (First) [ 30.100267] pcieport 0000:00:1d.0: AER: TLP Header: 34000000 08000052 00000000 00000000 [ 30.100372] thunderbolt 0000:0a:00.0: AER: can't recover (no error_detected callback) [ 30.100401] xhci_hcd 0000:3e:00.0: AER: can't recover (no error_detected callback) [ 30.100427] pcieport 0000:00:1d.0: AER: device recovery failed So disable AER service to avoid the noises from turning power rails on/off when the device is in low power states (D3hot and D3cold), as PCIe spec "5.2 Link State Power Management" states that TLP and DLLP transmission is disabled for a Link in L2/L3 Ready (D3hot), L2 (D3cold with aux power) and L3 (D3cold). Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=209149 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=215453 Fixes: 50310600ebda ("iommu/vt-d: Enable PCI ACS for platform opt in hint") Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com> --- v2: - Wording change. drivers/pci/pcie/aer.c | 31 +++++++++++++++++++++++++------ 1 file changed, 25 insertions(+), 6 deletions(-)