Skip to contentSkip to conversation

iommu-common: Do not try to deref a null iommu->lazy_flush() pointer when n < pool->hint

PATCH In review Sowmini Varadhan 2 participants 4 messages Sep 17, 2015 Library code

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 iommu-common: Do not try to deref a null iommu->lazy_flush() pointer when n < pool->hint

The check for invoking iommu->lazy_flush() from iommu_tbl_range_alloc() has to be refactored so that we only call ->lazy_flush() if it is non-null.

---
 lib/iommu-common.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)
lib/iommu-common.c +3 -3
index ff19f66..b1c93e9 100644
@@ -21,8 +21,7 @@ static	DEFINE_PER_CPU(unsigned int, iommu_hash_common);
 
 static inline bool need_flush(struct iommu_map_table *iommu)
 {
-	return (iommu->lazy_flush != NULL &&
-		(iommu->flags & IOMMU_NEED_FLUSH) != 0);
+	return ((iommu->flags & IOMMU_NEED_FLUSH) != 0);
 }
 
 static inline void set_flush(struct iommu_map_table *iommu)
@@ -211,7 +210,8 @@ unsigned long iommu_tbl_range_alloc(struct device *dev,
 			goto bail;
 		}
 	}
-	if (n < pool->hint || need_flush(iommu)) {
+	if (iommu->lazy_flush &&
+	    (n < pool->hint || need_flush(iommu))) {
 		clear_flush(iommu);
 		iommu->lazy_flush(iommu);
 	}
Signature · 1 line
1.7.1
maintainer
#1 lore raw
Quoted 2 lines · On Wed, 16 Sep 2015 21:50:43 -0400 Sowmini Varadhan…

The check for invoking iommu->lazy_flush() from iommu_tbl_range_alloc() has to be refactored so that we only call ->lazy_flush() if it is non-null.

Patch looks good.

But I don't know which kernel versions need the fix. So, when fixing a bug please always describe the end-user visible effects of that bug. Under what circumstances did your kernel crash?

#2 lore raw
Quoted 6 lines · Andrew Morton

Patch looks good.

But I don't know which kernel versions need the fix. So, when fixing a bug please always describe the end-user visible effects of that bug. Under what circumstances did your kernel crash?

I had a sparc kernel that was crashing when I was trying to process some very large perf.data files- the crash happens when the scsi driver calls into dma_4v_map_sg and thus the iommu_tbl_range_alloc().

Do you need me to respin the patch with the above info in the commit id?

--Sowmini

maintainer
#3 lore raw
Quoted 11 lines · On Thu, 17 Sep 2015 17:32:13 -0400 Sowmini Varadhan…
Quoted 6 lines · Andrew Morton

Patch looks good.

But I don't know which kernel versions need the fix. So, when fixing a bug please always describe the end-user visible effects of that bug. Under what circumstances did your kernel crash?

I had a sparc kernel that was crashing when I was trying to process some very large perf.data files- the crash happens when the scsi driver calls into dma_4v_map_sg and thus the iommu_tbl_range_alloc().

Cool, thanks.

Quoted 2 lines

Do you need me to respin the patch with the above info in the commit id?

No, that's fine. I updated the changelog, added cc:stable and shall send it upstream next week.