Skip to contentSkip to conversation

eeh: Fixing a bug when pci structure is null

PATCH In review Breno Leitao 5 participants 8 messages Feb 2, 2010 – Feb 24, 2010 1 message not in the archive

Subsystems (2)
lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw

PATCH 1/1 eeh: Fixing a bug when pci structure is null

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) and the kernel will crash with the following message:

 Unable to handle kernel paging request for data at address 0x000000a0
 Faulting instruction address: 0xc00000000006b8b4
 Oops: Kernel access of bad area, sig: 11 [#1]

NIP [c00000000006b8b4] .eeh_event_handler+0x10c/0x1a0 LR [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0

 Call Trace:
 [c0000003a80dff00] [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0
 [c0000003a80dff90] [c000000000031f1c] .kernel_thread+0x54/0x70

The bug occurs because pci_name() tries to access a null pointer. This patch just guarantee that pci_name() is not called on Null pointers.

---

  arch/powerpc/include/asm/ppc-pci.h          |    7 +++++++
  arch/powerpc/platforms/pseries/eeh.c        |    4 ++--
  arch/powerpc/platforms/pseries/eeh_driver.c |    4 ++--
  arch/powerpc/platforms/pseries/eeh_event.c  |    2 +-
  4 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/include/asm/ppc-pci.h b/arch/powerpc/include/asm/ppc-pci.h index 2828f9d..724dbe2 100644
--- a/arch/powerpc/include/asm/ppc-pci.h
+++ b/arch/powerpc/include/asm/ppc-pci.h
@@ -137,6 +137,13 @@ struct device_node * find_device_pe(struct device_node *dn);
void eeh_sysfs_add_device(struct pci_dev *pdev);
void eeh_sysfs_remove_device(struct pci_dev *pdev);

+static inline const char *eeh_pci_name(struct pci_dev *pdev) +{
+ if (NULL==pdev)
+ return "<null>";
+ return pci_name(pdev);
+}
+
#endif /* CONFIG_EEH */

#else /* CONFIG_PCI */
diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index ccd8dd0..f9360fe 100644
--- a/arch/powerpc/platforms/pseries/eeh.c
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -491,7 +491,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
pdn->eeh_mode & EEH_MODE_NOCHECK) {
ignored_check++;
pr_debug("EEH: Ignored check (%x) for %s %s\n",
- pdn->eeh_mode, pci_name (dev), dn->full_name);
+ pdn->eeh_mode, eeh_pci_name (dev), dn->full_name);
return 0;
}

@@ -515,7 +515,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
printk (KERN_ERR "EEH: %d reads ignored for recovering device at "
"location=%s driver=%s pci addr=%s\n", pdn->eeh_check_count, location,
- dev->driver->name, pci_name(dev));
+ dev->driver->name, eeh_pci_name(dev));
printk (KERN_ERR "EEH: Might be infinite loop in %s driver\n",
dev->driver->name);
dump_stack();
diff --git a/arch/powerpc/platforms/pseries/eeh_driver.c b/arch/powerpc/platforms/pseries/eeh_driver.c index ef8e454..8f948a0 100644
--- a/arch/powerpc/platforms/pseries/eeh_driver.c
+++ b/arch/powerpc/platforms/pseries/eeh_driver.c
@@ -337,7 +337,7 @@ struct pci_dn * handle_eeh_events (struct eeh_event *event)
location = location ? location : "unknown";
printk(KERN_ERR "EEH: Error: Cannot find partition endpoint "
"for location=%s pci addr=%s\n",
- location, pci_name(event->dev));
+ location, eeh_pci_name(event->dev));
return NULL;
}

@@ -368,7 +368,7 @@ struct pci_dn * handle_eeh_events (struct eeh_event *event)
pci_str = pci_name (frozen_pdn->pcidev);
drv_str = pcid_name (frozen_pdn->pcidev);
} else {
- pci_str = pci_name (event->dev);
+ pci_str = eeh_pci_name (event->dev);
drv_str = pcid_name (event->dev);
}

diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c index ddb80f5..ec5df8f 100644
--- a/arch/powerpc/platforms/pseries/eeh_event.c
+++ b/arch/powerpc/platforms/pseries/eeh_event.c
@@ -80,7 +80,7 @@ static int eeh_event_handler(void * dummy)
eeh_mark_slot(event->dn, EEH_MODE_RECOVERING);

printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n",
- pci_name(event->dev));
+ eeh_pci_name(event->dev));

  	pdn = handle_eeh_events(event);
#2 lore raw
Quoted 35 lines · Breno Leitao

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) and the kernel will crash with the following message:

 Unable to handle kernel paging request for data at address 0x000000a0
 Faulting instruction address: 0xc00000000006b8b4
 Oops: Kernel access of bad area, sig: 11 [#1]

NIP [c00000000006b8b4] .eeh_event_handler+0x10c/0x1a0 LR [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0

 Call Trace:
 [c0000003a80dff00] [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0
 [c0000003a80dff90] [c000000000031f1c] .kernel_thread+0x54/0x70

The bug occurs because pci_name() tries to access a null pointer. This patch just guarantee that pci_name() is not called on Null pointers.

---

  arch/powerpc/include/asm/ppc-pci.h          |    7 +++++++
  arch/powerpc/platforms/pseries/eeh.c        |    4 ++--
  arch/powerpc/platforms/pseries/eeh_driver.c |    4 ++--
  arch/powerpc/platforms/pseries/eeh_event.c  |    2 +-
  4 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/include/asm/ppc-pci.h b/arch/powerpc/include/asm/ppc-pci.h index 2828f9d..724dbe2 100644
--- a/arch/powerpc/include/asm/ppc-pci.h
+++ b/arch/powerpc/include/asm/ppc-pci.h
@@ -137,6 +137,13 @@ struct device_node * find_device_pe(struct device_node *dn);
void eeh_sysfs_add_device(struct pci_dev *pdev);
void eeh_sysfs_remove_device(struct pci_dev *pdev);

Quoted 5 lines

+static inline const char *eeh_pci_name(struct pci_dev *pdev) +{
+ if (NULL==pdev)
+ return "<null>";
+ return pci_name(pdev);

What about:

	return pdev ? pci_name(pdev) : "<null>";
Quoted 4 lines

+}
+
#endif /* CONFIG_EEH */

Quoted 11 lines

#else /* CONFIG_PCI */
diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index ccd8dd0..f9360fe 100644
--- a/arch/powerpc/platforms/pseries/eeh.c
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -491,7 +491,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
pdn->eeh_mode & EEH_MODE_NOCHECK) {
ignored_check++;
pr_debug("EEH: Ignored check (%x) for %s %s\n",
- pdn->eeh_mode, pci_name (dev), dn->full_name);
+ pdn->eeh_mode, eeh_pci_name (dev), dn->full_name);

No space after function name, please.

Quoted 9 lines
  		return 0;
  	}

@@ -515,7 +515,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
printk (KERN_ERR "EEH: %d reads ignored for recovering device at "
"location=%s driver=%s pci addr=%s\n", pdn->eeh_check_count, location,
- dev->driver->name, pci_name(dev));
+ dev->driver->name, eeh_pci_name(dev));

ditto

Quoted 16 lines

printk (KERN_ERR "EEH: Might be infinite loop in %s driver\n",
dev->driver->name);
dump_stack();
diff --git a/arch/powerpc/platforms/pseries/eeh_driver.c b/arch/powerpc/platforms/pseries/eeh_driver.c index ef8e454..8f948a0 100644
--- a/arch/powerpc/platforms/pseries/eeh_driver.c
+++ b/arch/powerpc/platforms/pseries/eeh_driver.c
@@ -337,7 +337,7 @@ struct pci_dn * handle_eeh_events (struct eeh_event *event)
location = location ? location : "unknown";
printk(KERN_ERR "EEH: Error: Cannot find partition endpoint "
"for location=%s pci addr=%s\n",
- location, pci_name(event->dev));
+ location, eeh_pci_name(event->dev));
return NULL;
}

Quoted 6 lines

@@ -368,7 +368,7 @@ struct pci_dn * handle_eeh_events (struct eeh_event *event)
pci_str = pci_name (frozen_pdn->pcidev);
drv_str = pcid_name (frozen_pdn->pcidev);
} else {
- pci_str = pci_name (event->dev);
+ pci_str = eeh_pci_name (event->dev);

ditto

Quoted 19 lines
  		drv_str = pcid_name (event->dev);
  	}

diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c index ddb80f5..ec5df8f 100644
--- a/arch/powerpc/platforms/pseries/eeh_event.c
+++ b/arch/powerpc/platforms/pseries/eeh_event.c
@@ -80,7 +80,7 @@ static int eeh_event_handler(void * dummy)
eeh_mark_slot(event->dn, EEH_MODE_RECOVERING);

printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n",
- pci_name(event->dev));
+ eeh_pci_name(event->dev));

pdn = handle_eeh_events(event);
_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://lists.ozlabs.org/listinfo/linuxppc-dev

Signature · 2 lines
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
#3 lore raw

PATCH 1/1 eeh: Fixing a bug when pci structure is null

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) the kernel will crash with the following message:

Unable to handle kernel paging request for data at address 0x000000a0
Faulting instruction address: 0xc00000000006b8b4
Oops: Kernel access of bad area, sig: 11 [#1]

NIP [c00000000006b8b4] .eeh_event_handler+0x10c/0x1a0 LR [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0

Call Trace:
[c0000003a80dff00] [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0
[c0000003a80dff90] [c000000000031f1c] .kernel_thread+0x54/0x70

The bug occurs because pci_name() tries to access a null pointer. This patch just guarantee that pci_name() is not called on Null pointers.

---
 arch/powerpc/include/asm/ppc-pci.h          |    5 +++++
 arch/powerpc/platforms/pseries/eeh.c        |    4 ++--
 arch/powerpc/platforms/pseries/eeh_driver.c |    4 ++--
 arch/powerpc/platforms/pseries/eeh_event.c  |    2 +-
 4 files changed, 10 insertions(+), 5 deletions(-)
arch/powerpc/include/asm/ppc-pci.h +5 -0
index 2828f9d..42fdff0 100644
@@ -137,6 +137,11 @@ struct device_node * find_device_pe(struct device_node *dn);
 void eeh_sysfs_add_device(struct pci_dev *pdev);
 void eeh_sysfs_remove_device(struct pci_dev *pdev);
 
+static inline const char *eeh_pci_name(struct pci_dev *pdev) 
+{ 
+	return pdev ? pci_name(pdev) : "<null>";
+} 
+
 #endif /* CONFIG_EEH */
 
 #else /* CONFIG_PCI */
arch/powerpc/platforms/pseries/eeh.c +2 -2
index ccd8dd0..3304f32 100644
@@ -491,7 +491,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
 	    pdn->eeh_mode & EEH_MODE_NOCHECK) {
 		ignored_check++;
 		pr_debug("EEH: Ignored check (%x) for %s %s\n",
-			 pdn->eeh_mode, pci_name (dev), dn->full_name);
+			 pdn->eeh_mode, eeh_pci_name(dev), dn->full_name);
 		return 0;
 	}
 
@@ -515,7 +515,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
 			printk (KERN_ERR "EEH: %d reads ignored for recovering device at "
 				"location=%s driver=%s pci addr=%s\n",
 				pdn->eeh_check_count, location,
-				dev->driver->name, pci_name(dev));
+				dev->driver->name, eeh_pci_name(dev));
 			printk (KERN_ERR "EEH: Might be infinite loop in %s driver\n",
 				dev->driver->name);
 			dump_stack();
arch/powerpc/platforms/pseries/eeh_driver.c +2 -2
index ef8e454..977d87d 100644
@@ -337,7 +337,7 @@ struct pci_dn * handle_eeh_events (struct eeh_event *event)
 		location = location ? location : "unknown";
 		printk(KERN_ERR "EEH: Error: Cannot find partition endpoint "
 		                "for location=%s pci addr=%s\n",
-		        location, pci_name(event->dev));
+		        location, eeh_pci_name(event->dev));
 		return NULL;
 	}
 
@@ -368,7 +368,7 @@ struct pci_dn * handle_eeh_events (struct eeh_event *event)
 		pci_str = pci_name (frozen_pdn->pcidev);
 		drv_str = pcid_name (frozen_pdn->pcidev);
 	} else {
-		pci_str = pci_name (event->dev);
+		pci_str = eeh_pci_name(event->dev);
 		drv_str = pcid_name (event->dev);
 	}
 	
arch/powerpc/platforms/pseries/eeh_event.c +1 -1
index ddb80f5..ec5df8f 100644
@@ -80,7 +80,7 @@ static int eeh_event_handler(void * dummy)
 	eeh_mark_slot(event->dn, EEH_MODE_RECOVERING);
 
 	printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n",
-	       pci_name(event->dev));
+	       eeh_pci_name(event->dev));
 
 	pdn = handle_eeh_events(event);
 
Signature · 1 line
1.6.0.2
#4 lore raw

Hi Ben,

I'd like to ask about this patch ? Should I re-submit ?

Thanks,

Quoted 4 lines · Breno Leitao

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) the kernel will crash with the following message:

#5 lore raw

Hi Paul, Breno,

Some confusion -- I've been out of the loop for a while -- I assume its still Paul who is pushing
these patches upstream, and not Ben? So Breno, maybe you should resend the patch to Paul?

--linas

Quoted 12 lines · Breno Leitao

Hi Ben,

I'd like to ask about this patch ? Should I re-submit ?

Thanks,

Quoted 4 lines · Breno Leitao

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) the kernel will crash with the following message:

#6 lore raw
Quoted 5 lines · Linas Vepstas

Some confusion -- I've been out of the loop for a while -- I assume its still Paul who is pushing
these patches upstream, and not Ben? So Breno, maybe you should resend the patch to Paul?

No, it's me.

Cheers,
Ben.

#7 lore raw
In reply to Breno Leitao
Quoted 11 lines · Breno Leitao

Hi Ben,

I'd like to ask about this patch ? Should I re-submit ?

Thanks,

Quoted 4 lines · Breno Leitao

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) the kernel will crash with the following message:

It should be in -next, can you dbl check ?

Cheers,
Ben.

#8 lore raw
Quoted 7 lines · Benjamin Herrenschmidt
Quoted 6 lines · Breno Leitao

Hi Ben,

I'd like to ask about this patch ? Should I re-submit ?

Thanks,

Quoted 7 lines
Quoted 5 lines
Quoted 4 lines · Breno Leitao

During a EEH recover, the pci_dev structure can be null, mainly if an eeh event is detected during cpi config operation. In this case, the pci_dev will not be known (and will be null) the kernel will crash with the following message:

It should be in -next, can you dbl check ?

I just confirmed the patch is in the -next tree.

Mike

Quoted 9 lines

Cheers,
Ben.

_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://lists.ozlabs.org/listinfo/linuxppc-dev