[PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

The dma_direct_supported() function intends to check the DMA mask against
specific values. However, the phys_to_dma() function includes the SME
encryption mask, which defeats the intended purpose of the check. This
results in drivers that support less than 48-bit DMA (SME encryption mask
is bit 47) from being able to set the DMA mask successfully when SME is
active, which results in the driver failing to initialize.

Change the function used to check the mask from phys_to_dma() to
__phys_to_dma() so that the SME encryption mask is not part of the check.

Fixes: c1d0af1a1d5d ("kernel/dma/direct: take DMA offset into account in dma_direct_supported")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 kernel/dma/direct.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
index 22a12ab..375c77e 100644
--- a/kernel/dma/direct.c
+++ b/kernel/dma/direct.c
@@ -309,7 +309,12 @@ int dma_direct_supported(struct device *dev, u64 mask)
 
 	min_mask = min_t(u64, min_mask, (max_pfn - 1) << PAGE_SHIFT);
 
-	return mask >= phys_to_dma(dev, min_mask);
+	/*
+	 * This check needs to be against the actual bit mask value, so
+	 * use __phys_to_dma() here so that the SME encryption mask isn't
+	 * part of the check.
+	 */
+	return mask >= __phys_to_dma(dev, min_mask);
 }
 
 int dma_direct_mapping_error(struct device *dev, dma_addr_t dma_addr)

[PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

The dma_direct_supported() function intends to check the DMA mask against
specific values. However, the phys_to_dma() function includes the SME
encryption mask, which defeats the intended purpose of the check. This
results in drivers that support less than 48-bit DMA (SME encryption mask
is bit 47) from being able to set the DMA mask successfully when SME is
active, which results in the driver failing to initialize.

Change the function used to check the mask from phys_to_dma() to
__phys_to_dma() so that the SME encryption mask is not part of the check.

Fixes: c1d0af1a1d5d ("kernel/dma/direct: take DMA offset into account in dma_direct_supported")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 kernel/dma/direct.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
index 22a12ab..375c77e 100644
--- a/kernel/dma/direct.c
+++ b/kernel/dma/direct.c
@@ -309,7 +309,12 @@ int dma_direct_supported(struct device *dev, u64 mask)
 
 	min_mask = min_t(u64, min_mask, (max_pfn - 1) << PAGE_SHIFT);
 
-	return mask >= phys_to_dma(dev, min_mask);
+	/*
+	 * This check needs to be against the actual bit mask value, so
+	 * use __phys_to_dma() here so that the SME encryption mask isn't
+	 * part of the check.
+	 */
+	return mask >= __phys_to_dma(dev, min_mask);
 }
 
 int dma_direct_mapping_error(struct device *dev, dma_addr_t dma_addr)

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Christoph Hellwig]

The mail seems to be so oddly encoded so that git-am fails on it.  Can
you resend as plain text?

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Christoph Hellwig]

The mail seems to be so oddly encoded so that git-am fails on it.  Can
you resend as plain text?

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

On 12/15/2018 04:55 AM, Christoph Hellwig wrote:
> The mail seems to be so oddly encoded so that git-am fails on it.  Can
> you resend as plain text?

Hmmm... not sure what happened with that, but yeah, looking at the message
source shows something strange went on. Let me take a look and I'll try to
get a good version to you tommorow (Monday).

Thanks,
Tom

> 

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

On 12/15/2018 04:55 AM, Christoph Hellwig wrote:
> The mail seems to be so oddly encoded so that git-am fails on it.  Can
> you resend as plain text?

Hmmm... not sure what happened with that, but yeah, looking at the message
source shows something strange went on. Let me take a look and I'll try to
get a good version to you tommorow (Monday).

Thanks,
Tom

> 

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

On 12/15/2018 04:55 AM, Christoph Hellwig wrote:
> The mail seems to be so oddly encoded so that git-am fails on it.  Can
> you resend as plain text?

Hmmm... not sure what happened with that, but yeah, looking at the message
source shows something strange went on. Let me take a look and I'll try to
get a good version to you tommorow (Monday).

Thanks,
Tom

> 

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

On 12/16/2018 05:41 PM, Tom Lendacky wrote:
> On 12/15/2018 04:55 AM, Christoph Hellwig wrote:
>> The mail seems to be so oddly encoded so that git-am fails on it.  Can
>> you resend as plain text?
> 
> Hmmm... not sure what happened with that, but yeah, looking at the message
> source shows something strange went on. Let me take a look and I'll try to
> get a good version to you tommorow (Monday).

Must have been something with stgit... just resent using git format-patch
and git send-email and it looks ok.  Let me know if it's still not right
when it gets to you.

Thanks,
Tom

> 
> Thanks,
> Tom
> 
>>

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

On 12/16/2018 05:41 PM, Tom Lendacky wrote:
> On 12/15/2018 04:55 AM, Christoph Hellwig wrote:
>> The mail seems to be so oddly encoded so that git-am fails on it.  Can
>> you resend as plain text?
> 
> Hmmm... not sure what happened with that, but yeah, looking at the message
> source shows something strange went on. Let me take a look and I'll try to
> get a good version to you tommorow (Monday).

Must have been something with stgit... just resent using git format-patch
and git send-email and it looks ok.  Let me know if it's still not right
when it gets to you.

Thanks,
Tom

> 
> Thanks,
> Tom
> 
>>

[PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Lendacky, Thomas]

commit c92a54cfa0257e8ffd66b2a17d49e9c0bd4b769f upstream

This fix appears in 4.20, but dma_direct_supported() was changed in 4.20
such that the original version of the fix will not apply to previous
versions of the kernel. The fix only applies to the 4.19-stable tree and
has been backported for that tree.

The dma_direct_supported() function intends to check the DMA mask against
specific values. However, the phys_to_dma() function includes the SME
encryption mask, which defeats the intended purpose of the check. This
results in drivers that support less than 48-bit DMA (SME encryption mask
is bit 47) from being able to set the DMA mask successfully when SME is
active, which results in the driver failing to initialize.

Change the function used to check the mask from phys_to_dma() to
__phys_to_dma() so that the SME encryption mask is not part of the check.

Fixes: c1d0af1a1d5d ("kernel/dma/direct: take DMA offset into account in dma_direct_supported")
Cc: <stable@vger.kernel.org> # 4.19.x
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 kernel/dma/direct.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
index de87b02..1d2f147 100644
--- a/kernel/dma/direct.c
+++ b/kernel/dma/direct.c
@@ -168,7 +168,12 @@ int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl, int nents,
 int dma_direct_supported(struct device *dev, u64 mask)
 {
 #ifdef CONFIG_ZONE_DMA
-	if (mask < phys_to_dma(dev, DMA_BIT_MASK(ARCH_ZONE_DMA_BITS)))
+	/*
+	 * This check needs to be against the actual bit mask value, so
+	 * use __phys_to_dma() here so that the SME encryption mask isn't
+	 * part of the check.
+	 */
+	if (mask < __phys_to_dma(dev, DMA_BIT_MASK(ARCH_ZONE_DMA_BITS)))
 		return 0;
 #else
 	/*
@@ -176,8 +181,12 @@ int dma_direct_supported(struct device *dev, u64 mask)
 	 * to be able to satisfy them - either by not supporting more physical
 	 * memory, or by providing a ZONE_DMA32.  If neither is the case, the
 	 * architecture needs to use an IOMMU instead of the direct mapping.
+	 *
+	 * This check needs to be against the actual bit mask value, so
+	 * use __phys_to_dma() here so that the SME encryption mask isn't
+	 * part of the check.
 	 */
-	if (mask < phys_to_dma(dev, DMA_BIT_MASK(32)))
+	if (mask < __phys_to_dma(dev, DMA_BIT_MASK(32)))
 		return 0;
 #endif
 	/*
-- 
1.9.1

Re: [PATCH] kernel/dma/direct: Do not include SME mask in the DMA supported check

[Greg KH]

On Sat, Dec 22, 2018 at 04:32:40PM +0000, Lendacky, Thomas wrote:
> commit c92a54cfa0257e8ffd66b2a17d49e9c0bd4b769f upstream
> 
> This fix appears in 4.20, but dma_direct_supported() was changed in 4.20
> such that the original version of the fix will not apply to previous
> versions of the kernel. The fix only applies to the 4.19-stable tree and
> has been backported for that tree.
> 
> The dma_direct_supported() function intends to check the DMA mask against
> specific values. However, the phys_to_dma() function includes the SME
> encryption mask, which defeats the intended purpose of the check. This
> results in drivers that support less than 48-bit DMA (SME encryption mask
> is bit 47) from being able to set the DMA mask successfully when SME is
> active, which results in the driver failing to initialize.
> 
> Change the function used to check the mask from phys_to_dma() to
> __phys_to_dma() so that the SME encryption mask is not part of the check.
> 
> Fixes: c1d0af1a1d5d ("kernel/dma/direct: take DMA offset into account in dma_direct_supported")
> Cc: <stable@vger.kernel.org> # 4.19.x
> Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
> ---
>  kernel/dma/direct.c | 13 +++++++++++--
>  1 file changed, 11 insertions(+), 2 deletions(-)

Now queued up, thanks.

greg k-h