Skip to contentSkip to conversation

powernv/eeh: Fix oops when probing cxl devices

RFC PATCH In review Frederic Barrat 2 participants 4 messages Sep 27, 2019 – Oct 16, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/1 powernv/eeh: Fix oops when probing cxl devices

Recent cleanup in the way EEH support is added to a device causes a kernel oops when the cxl driver probes a device and creates virtual devices discovered on the FPGA:

    BUG: Kernel NULL pointer dereference at 0x000000a0
    Faulting instruction address: 0xc000000000048070
    Oops: Kernel access of bad area, sig: 7 [#1]
    ...
    NIP [c000000000048070] eeh_add_device_late.part.9+0x50/0x1e0
    LR [c00000000004805c] eeh_add_device_late.part.9+0x3c/0x1e0
    Call Trace:
    [c000200e43983900] [c00000000079e250] _dev_info+0x5c/0x6c (unreliable)
    [c000200e43983980] [c0000000000d1ad0] pnv_pcibios_bus_add_device+0x60/0xb0
    [c000200e439839f0] [c0000000000606d0] pcibios_bus_add_device+0x40/0x60
    [c000200e43983a10] [c0000000006aa3a0] pci_bus_add_device+0x30/0x100
    [c000200e43983a80] [c0000000006aa4d4] pci_bus_add_devices+0x64/0xd0
    [c000200e43983ac0] [c00800001c429118] cxl_pci_vphb_add+0xe0/0x130 [cxl]
    [c000200e43983b00] [c00800001c4242ac] cxl_probe+0x504/0x5b0 [cxl]
    [c000200e43983bb0] [c0000000006bba1c] local_pci_probe+0x6c/0x110
    [c000200e43983c30] [c000000000159278] work_for_cpu_fn+0x38/0x60

The root cause is that those cxl virtual devices don't have a representation in the device tree and therefore no associated pci_dn structure. In eeh_add_device_late(), pdn is NULL, so edev is NULL and we oops.

We never had explicit support for EEH for those virtual devices. Instead, EEH events are reported to the (real) pci device and handled by the cxl driver. Which can then forward to the virtual devices and handle dependencies. The fact that we try adding EEH support for the virtual devices is new and a side-effect of the recent cleanup.

This patch fixes it by skipping adding EEH support on powernv for devices which don't have a pci_dn structure.

Fixes b905f8cdca77 ("powerpc/eeh: EEH for pSeries hot plug")

---

Sending as an RFC, as I'm afraid of hiding potential issues and would be interested in comments. The powernv eeh code expects a struct pci_dn, so the fix seems safe. I'm wondering if there could be cases (other than capi virtual devices) where we'd want to blow up and fix instead of going undetected with this patch.

 arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/powernv/eeh-powernv.c +1 -1
index 6bc24a47e9ef..6f300ab7f0e9 100644
@@ -42,7 +42,7 @@ void pnv_pcibios_bus_add_device(struct pci_dev *pdev)
 {
 	struct pci_dn *pdn = pci_get_pdn(pdev);
 
-	if (eeh_has_flag(EEH_FORCE_DISABLED))
+	if (!pdn || eeh_has_flag(EEH_FORCE_DISABLED))
 		return;
 
 	dev_dbg(&pdev->dev, "EEH: Setting up device\n");
Signature · 1 line
2.21.0
#1 lore raw
Quoted 38 lines · Frederic Barrat

Recent cleanup in the way EEH support is added to a device causes a kernel oops when the cxl driver probes a device and creates virtual devices discovered on the FPGA:

    BUG: Kernel NULL pointer dereference at 0x000000a0
    Faulting instruction address: 0xc000000000048070
    Oops: Kernel access of bad area, sig: 7 [#1]
    ...
    NIP [c000000000048070] eeh_add_device_late.part.9+0x50/0x1e0
    LR [c00000000004805c] eeh_add_device_late.part.9+0x3c/0x1e0
    Call Trace:
    [c000200e43983900] [c00000000079e250] _dev_info+0x5c/0x6c (unreliable)
    [c000200e43983980] [c0000000000d1ad0] pnv_pcibios_bus_add_device+0x60/0xb0
    [c000200e439839f0] [c0000000000606d0] pcibios_bus_add_device+0x40/0x60
    [c000200e43983a10] [c0000000006aa3a0] pci_bus_add_device+0x30/0x100
    [c000200e43983a80] [c0000000006aa4d4] pci_bus_add_devices+0x64/0xd0
    [c000200e43983ac0] [c00800001c429118] cxl_pci_vphb_add+0xe0/0x130 [cxl]
    [c000200e43983b00] [c00800001c4242ac] cxl_probe+0x504/0x5b0 [cxl]
    [c000200e43983bb0] [c0000000006bba1c] local_pci_probe+0x6c/0x110
    [c000200e43983c30] [c000000000159278] work_for_cpu_fn+0x38/0x60

The root cause is that those cxl virtual devices don't have a representation in the device tree and therefore no associated pci_dn structure. In eeh_add_device_late(), pdn is NULL, so edev is NULL and we oops.

We never had explicit support for EEH for those virtual devices. Instead, EEH events are reported to the (real) pci device and handled by the cxl driver. Which can then forward to the virtual devices and handle dependencies. The fact that we try adding EEH support for the virtual devices is new and a side-effect of the recent cleanup.

This patch fixes it by skipping adding EEH support on powernv for devices which don't have a pci_dn structure.

Fixes b905f8cdca77 ("powerpc/eeh: EEH for pSeries hot plug")
Quoted 7 lines

---

Sending as an RFC, as I'm afraid of hiding potential issues and would be interested in comments. The powernv eeh code expects a struct pci_dn, so the fix seems safe. I'm wondering if there could be cases (other than capi virtual devices) where we'd want to blow up and fix instead of going undetected with this patch.

Looks good to me.

I do think it would be good to detect a missing pci_dn (WARN_ONCE() might be appropriate). However to implement it, pnv_pcibios_bus_add_device() would need a way to detect that a struct pci_dev is a cxl virtual device. I don't see an easy way to do that; do you know if it's possible?

One last thing: pseries_pcibios_bus_add_device() also requires a pci_dn. Do you know if it's possible to trigger a similar issue there, or is it not possible for some reason?

Cheers,
Sam.

Quoted 19 lines
 arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/powernv/eeh-powernv.c +1 -1
index 6bc24a47e9ef..6f300ab7f0e9 100644
@@ -42,7 +42,7 @@ void pnv_pcibios_bus_add_device(struct pci_dev *pdev)
 {
 	struct pci_dn *pdn = pci_get_pdn(pdev);
 
-	if (eeh_has_flag(EEH_FORCE_DISABLED))
+	if (!pdn || eeh_has_flag(EEH_FORCE_DISABLED))
 		return;
 
 	dev_dbg(&pdev->dev, "EEH: Setting up device\n");
Signature · 1 line
2.21.0
#2 lore raw

Le 15/10/2019 à 07:42, Sam Bobroff a écrit :

Quoted 47 lines · Frederic Barrat
Quoted 38 lines

Recent cleanup in the way EEH support is added to a device causes a kernel oops when the cxl driver probes a device and creates virtual devices discovered on the FPGA:

     BUG: Kernel NULL pointer dereference at 0x000000a0
     Faulting instruction address: 0xc000000000048070
     Oops: Kernel access of bad area, sig: 7 [#1]
     ...
     NIP [c000000000048070] eeh_add_device_late.part.9+0x50/0x1e0
     LR [c00000000004805c] eeh_add_device_late.part.9+0x3c/0x1e0
     Call Trace:
     [c000200e43983900] [c00000000079e250] _dev_info+0x5c/0x6c (unreliable)
     [c000200e43983980] [c0000000000d1ad0] pnv_pcibios_bus_add_device+0x60/0xb0
     [c000200e439839f0] [c0000000000606d0] pcibios_bus_add_device+0x40/0x60
     [c000200e43983a10] [c0000000006aa3a0] pci_bus_add_device+0x30/0x100
     [c000200e43983a80] [c0000000006aa4d4] pci_bus_add_devices+0x64/0xd0
     [c000200e43983ac0] [c00800001c429118] cxl_pci_vphb_add+0xe0/0x130 [cxl]
     [c000200e43983b00] [c00800001c4242ac] cxl_probe+0x504/0x5b0 [cxl]
     [c000200e43983bb0] [c0000000006bba1c] local_pci_probe+0x6c/0x110
     [c000200e43983c30] [c000000000159278] work_for_cpu_fn+0x38/0x60

The root cause is that those cxl virtual devices don't have a representation in the device tree and therefore no associated pci_dn structure. In eeh_add_device_late(), pdn is NULL, so edev is NULL and we oops.

We never had explicit support for EEH for those virtual devices. Instead, EEH events are reported to the (real) pci device and handled by the cxl driver. Which can then forward to the virtual devices and handle dependencies. The fact that we try adding EEH support for the virtual devices is new and a side-effect of the recent cleanup.

This patch fixes it by skipping adding EEH support on powernv for devices which don't have a pci_dn structure.

Fixes b905f8cdca77 ("powerpc/eeh: EEH for pSeries hot plug")
Quoted 7 lines

---

Sending as an RFC, as I'm afraid of hiding potential issues and would be interested in comments. The powernv eeh code expects a struct pci_dn, so the fix seems safe. I'm wondering if there could be cases (other than capi virtual devices) where we'd want to blow up and fix instead of going undetected with this patch.

Quoted 7 lines

Looks good to me.

I do think it would be good to detect a missing pci_dn (WARN_ONCE() might be appropriate). However to implement it, pnv_pcibios_bus_add_device() would need a way to detect that a struct pci_dev is a cxl virtual device. I don't see an easy way to do that; do you know if it's possible?

I think I found a solution. There's a cxl_pci_is_vphb_device() which is fairly cheap and would do the job. Sorry, I didn't think about it at first.

Quoted 3 lines

One last thing: pseries_pcibios_bus_add_device() also requires a pci_dn. Do you know if it's possible to trigger a similar issue there, or is it not possible for some reason?

I don't think anybody is using capi in a lpar, but it should theorically be possible to hit it. I'll dig some more tomorrow and adjust the patch when resubmitting.

Thanks!

Fred

Quoted 23 lines

Cheers,
Sam.

Quoted 19 lines
  arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/powernv/eeh-powernv.c +1 -1
index 6bc24a47e9ef..6f300ab7f0e9 100644
@@ -42,7 +42,7 @@ void pnv_pcibios_bus_add_device(struct pci_dev *pdev)
  {
  	struct pci_dn *pdn = pci_get_pdn(pdev);
 
-	if (eeh_has_flag(EEH_FORCE_DISABLED))
+	if (!pdn || eeh_has_flag(EEH_FORCE_DISABLED))
  		return;
 
  	dev_dbg(&pdev->dev, "EEH: Setting up device\n");
Signature · 1 line
2.21.0
#3 lore raw

Le 15/10/2019 à 21:41, Frederic Barrat a écrit :

Quoted 52 lines

Le 15/10/2019 à 07:42, Sam Bobroff a écrit :

Quoted 49 lines · Frederic Barrat
Quoted 38 lines

Recent cleanup in the way EEH support is added to a device causes a kernel oops when the cxl driver probes a device and creates virtual devices discovered on the FPGA:

     BUG: Kernel NULL pointer dereference at 0x000000a0
     Faulting instruction address: 0xc000000000048070
     Oops: Kernel access of bad area, sig: 7 [#1]
     ...
     NIP [c000000000048070] eeh_add_device_late.part.9+0x50/0x1e0
     LR [c00000000004805c] eeh_add_device_late.part.9+0x3c/0x1e0
     Call Trace:
     [c000200e43983900] [c00000000079e250] _dev_info+0x5c/0x6c (unreliable)
     [c000200e43983980] [c0000000000d1ad0] pnv_pcibios_bus_add_device+0x60/0xb0
     [c000200e439839f0] [c0000000000606d0] pcibios_bus_add_device+0x40/0x60
     [c000200e43983a10] [c0000000006aa3a0] pci_bus_add_device+0x30/0x100
     [c000200e43983a80] [c0000000006aa4d4] pci_bus_add_devices+0x64/0xd0
     [c000200e43983ac0] [c00800001c429118] cxl_pci_vphb_add+0xe0/0x130 [cxl]
     [c000200e43983b00] [c00800001c4242ac] cxl_probe+0x504/0x5b0 [cxl]
     [c000200e43983bb0] [c0000000006bba1c] local_pci_probe+0x6c/0x110
     [c000200e43983c30] [c000000000159278] work_for_cpu_fn+0x38/0x60

The root cause is that those cxl virtual devices don't have a representation in the device tree and therefore no associated pci_dn structure. In eeh_add_device_late(), pdn is NULL, so edev is NULL and we oops.

We never had explicit support for EEH for those virtual devices. Instead, EEH events are reported to the (real) pci device and handled by the cxl driver. Which can then forward to the virtual devices and handle dependencies. The fact that we try adding EEH support for the virtual devices is new and a side-effect of the recent cleanup.

This patch fixes it by skipping adding EEH support on powernv for devices which don't have a pci_dn structure.

Fixes b905f8cdca77 ("powerpc/eeh: EEH for pSeries hot plug")
Quoted 7 lines

---

Sending as an RFC, as I'm afraid of hiding potential issues and would be interested in comments. The powernv eeh code expects a struct pci_dn, so the fix seems safe. I'm wondering if there could be cases (other than capi virtual devices) where we'd want to blow up and fix instead of going undetected with this patch.

Looks good to me.

Quoted 8 lines
Quoted 5 lines

I do think it would be good to detect a missing pci_dn (WARN_ONCE() might be appropriate).  However to implement it, pnv_pcibios_bus_add_device() would need a way to detect that a struct pci_dev is a cxl virtual device. I don't see an easy way to do that; do you know if it's possible?

I think I found a solution. There's a cxl_pci_is_vphb_device() which is fairly cheap and would do the job. Sorry, I didn't think about it at first.

Not sure what I was smoking... this is all internal to the cxl module and would require some changes to make it available to the kernel. Not impossible, but not trivial either. And I'm not convinced it's worth it to just be able to call WARN.

Fred

Quoted 35 lines
Quoted 4 lines

One last thing: pseries_pcibios_bus_add_device() also requires a pci_dn. Do you know if it's possible to trigger a similar issue there, or is it not possible for some reason?

I don't think anybody is using capi in a lpar, but it should theorically be possible to hit it. I'll dig some more tomorrow and adjust the patch when resubmitting.

Thanks!

  Fred

Quoted 21 lines

Cheers,
Sam.

Quoted 17 lines

  arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/powernv/eeh-powernv.c +1 -1
index 6bc24a47e9ef..6f300ab7f0e9 100644
@@ -42,7 +42,7 @@ void pnv_pcibios_bus_add_device(struct pci_dev *pdev)
   {
       struct pci_dn *pdn = pci_get_pdn(pdev);
-    if (eeh_has_flag(EEH_FORCE_DISABLED))
+    if (!pdn || eeh_has_flag(EEH_FORCE_DISABLED))
           return;
       dev_dbg(&pdev->dev, "EEH: Setting up device\n");
Signature · 1 line
2.21.0