dmaengine: do not allow access outside of unmap_pool
Quoted 29 lines · Colin King
Quoted 18 lines
From: Colin Ian King <[email protected]>
When
CONFIG_DMA_ENGINE_RAIDis defined, unmap_pool[] is just 1 element in size, however, allows orders of 2..8 to access outsideunmap_pooland returns an invalid address. Ensure we fall into the default path and report aBUG()whenCONFIG_DMA_ENGINE_RAIDis defined and order is out of range.Signed-off-by Colin Ian King <[email protected]>--- drivers/dma/dmaengine.c | 2 ++ 1 file changed, 2 insertions(+)drivers/dma/dmaengine.c +0
-0drivers/dma/dmaengine.c static struct dmaengine_unmap_pool *__get_unmap_pool(int nr) · line 1103
Quoted 4 lines
drivers/dma/dmaengine.c +1
-0@@ -1100,12 +1100,14 @@ static struct dmaengine_unmap_pool *__get_unmap_pool(int nr)switch (order) { case 0 ... 1:return &unmap_pool[0 ]; +#if IS_ENABLED (CONFIG_DMA_ENGINE_RAID )Okay if
CONFIG_DMA_ENGINE_RAIDis enabled (m or y) thenIS_ENABLEDreturn 1, so we will go inside and not fall into default. And I though by changelog that you want it to go to default inCONFIG_DMA_ENGINE_RAIDis defined!
Quoted 1 line
What did I miss...
Here it should be when CONFIG_DMA_ENGINE_RAID is NOT defined, unmap_pool[] is just 1 element in size, and the function "__get_unmap_pool" will access outside of the array unmap_pool[] in case orders of 2..8 and returns an invalid address, and I encountered the issue.
I think the patch is needed to avoid visiting outside of the array unmap_pool[] if CONFIG_DMA_ENGINE_RAID is NOT defined.
Thanks
Xulin
Quoted 13 lines
Quoted 13 lines
drivers/dma/dmaengine.c +1
-1case 2 ... 4:return &unmap_pool[1 ]; case 5 ... 7:return &unmap_pool[2 ];case 8 :return &unmap_pool[3 ]; +#endif default :BUG ();return NULL ; --2.8.1
Quoted 39 lines · Xulin Sun
Quoted 14 lines · Colin King
Quoted 13 lines
From: Colin Ian King <[email protected]>
When
CONFIG_DMA_ENGINE_RAIDis defined, unmap_pool[] is just 1 element in size, however, allows orders of 2..8 to access outsideunmap_pooland returns an invalid address. Ensure we fall into the default path and report aBUG()whenCONFIG_DMA_ENGINE_RAIDis defined and order is out of range.Signed-off-by Colin Ian King <[email protected]>--- drivers/dma/dmaengine.c | 2 ++ 1 file changed, 2 insertions(+)Quoted 5 lines
drivers/dma/dmaengine.c static struct dmaengine_unmap_pool
Quoted 5 lines
drivers/dma/dmaengine.c +0
-0@@ -1100,12 +1100,14 @@ static struct dmaengine_unmap_pool*__get_unmap_pool(int nr)
Quoted 11 lines
drivers/dma/dmaengine.c static struct dmaengine_unmap_pool · line 1103
Quoted 4 lines
drivers/dma/dmaengine.c +1
-0switch (order) { case 0 ... 1:return &unmap_pool[0 ]; +#if IS_ENABLED (CONFIG_DMA_ENGINE_RAID )Okay if
CONFIG_DMA_ENGINE_RAIDis enabled (m or y) thenIS_ENABLEDreturn 1, so we will go inside and not fall into default. And I though by changelog that you want it to go to default inCONFIG_DMA_ENGINE_RAIDis defined!What did I miss...
Here it should be when
CONFIG_DMA_ENGINE_RAIDis NOT defined, unmap_pool[] is just 1
element in size, and the function "__get_unmap_pool" will access outside of the array unmap_pool[]
in case orders of 2..8 and returns an invalid address, and I encountered the issue.
Quoted 2 lines
I think the patch is needed to avoid visiting outside of the array unmap_pool[] if
CONFIG_DMA_ENGINE_RAIDis NOT defined.
Exactly. Thanks for explaining, I missed the original query, apologies for missing that.
Colin
Quoted 16 lines
Thanks
XulinQuoted 13 lines
Quoted 13 lines
drivers/dma/dmaengine.c +1
-1case 2 ... 4:return &unmap_pool[1 ]; case 5 ... 7:return &unmap_pool[2 ];case 8 :return &unmap_pool[3 ]; +#endif default :BUG ();return NULL ; --2.8.1