Re: [PATCH] pci/quirks: ITE 8893 needs quirk_use_pcie_bridge_dma_alias

[Bjorn Helgaas]

On Mon, Apr 10, 2017 at 09:01:27PM -0400, Jarod Wilson wrote:
> This bridge has the same problems as the ITE 8892, which were resulting in
> crippling an older PCI 1Gbps NIC down to 45Mbps throughput with IOMMU and
> VT-d enabled. With the patch, this old e1000 goes back up to ~900Mbps.
> 
> Suggested-by: Alex Williamson 
> CC: linux-...@vger.kernel.org
> CC: Alex Williamson 
> CC: Bjorn Helgaas 
> Signed-off-by: Jarod Wilson 

Applied with Alex's reviewed-by to pci/iommu for v4.12, thanks!

> ---
>  drivers/pci/quirks.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 9236e40ac055..c95fbf2431f8 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -3914,6 +3914,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
>  DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
>  /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
>  DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
> +/* ITE 8893 has the same problem as the 8892 */
> +DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8893, quirk_use_pcie_bridge_dma_alias);
>  /* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
>  DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
>  
> -- 
> 2.12.1
> 

Re: [PATCH] pci/quirks: ITE 8893 needs quirk_use_pcie_bridge_dma_alias

[Bjorn Helgaas]

On Mon, Apr 10, 2017 at 09:01:27PM -0400, Jarod Wilson wrote:
> This bridge has the same problems as the ITE 8892, which were resulting in
> crippling an older PCI 1Gbps NIC down to 45Mbps throughput with IOMMU and
> VT-d enabled. With the patch, this old e1000 goes back up to ~900Mbps.
> 
> Suggested-by: Alex Williamson 
> CC: linux-...@vger.kernel.org
> CC: Alex Williamson 
> CC: Bjorn Helgaas 
> Signed-off-by: Jarod Wilson 

Applied with Alex's reviewed-by to pci/iommu for v4.12, thanks!

> ---
>  drivers/pci/quirks.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 9236e40ac055..c95fbf2431f8 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -3914,6 +3914,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
>  DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
>  /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
>  DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
> +/* ITE 8893 has the same problem as the 8892 */
> +DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8893, quirk_use_pcie_bridge_dma_alias);
>  /* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
>  DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
>  
> -- 
> 2.12.1
> 

Re: [PATCH] pci/quirks: ITE 8893 needs quirk_use_pcie_bridge_dma_alias

[Alex Williamson]

On Mon, 10 Apr 2017 21:01:27 -0400
Jarod Wilson  wrote:

> This bridge has the same problems as the ITE 8892, which were resulting in
> crippling an older PCI 1Gbps NIC down to 45Mbps throughput with IOMMU and
> VT-d enabled. With the patch, this old e1000 goes back up to ~900Mbps.
> 
> Suggested-by: Alex Williamson 
> CC: linux-...@vger.kernel.org
> CC: Alex Williamson 
> CC: Bjorn Helgaas 
> Signed-off-by: Jarod Wilson 
> ---
>  drivers/pci/quirks.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 9236e40ac055..c95fbf2431f8 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -3914,6 +3914,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
>  DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
>  /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
>  DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
> +/* ITE 8893 has the same problem as the 8892 */
> +DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8893, quirk_use_pcie_bridge_dma_alias);
>  /* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
>  DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);


Reviewed-by: Alex Williamson 

Re: [PATCH] pci/quirks: ITE 8893 needs quirk_use_pcie_bridge_dma_alias

[Alex Williamson]

On Mon, 10 Apr 2017 21:01:27 -0400
Jarod Wilson  wrote:

> This bridge has the same problems as the ITE 8892, which were resulting in
> crippling an older PCI 1Gbps NIC down to 45Mbps throughput with IOMMU and
> VT-d enabled. With the patch, this old e1000 goes back up to ~900Mbps.
> 
> Suggested-by: Alex Williamson 
> CC: linux-...@vger.kernel.org
> CC: Alex Williamson 
> CC: Bjorn Helgaas 
> Signed-off-by: Jarod Wilson 
> ---
>  drivers/pci/quirks.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 9236e40ac055..c95fbf2431f8 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -3914,6 +3914,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
>  DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
>  /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
>  DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
> +/* ITE 8893 has the same problem as the 8892 */
> +DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8893, quirk_use_pcie_bridge_dma_alias);
>  /* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
>  DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);


Reviewed-by: Alex Williamson 

[PATCH] pci/quirks: ITE 8893 needs quirk_use_pcie_bridge_dma_alias

[Jarod Wilson]

This bridge has the same problems as the ITE 8892, which were resulting in
crippling an older PCI 1Gbps NIC down to 45Mbps throughput with IOMMU and
VT-d enabled. With the patch, this old e1000 goes back up to ~900Mbps.

Suggested-by: Alex Williamson 
CC: linux-...@vger.kernel.org
CC: Alex Williamson 
CC: Bjorn Helgaas 
Signed-off-by: Jarod Wilson 
---
 drivers/pci/quirks.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 9236e40ac055..c95fbf2431f8 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -3914,6 +3914,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
 DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
 /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
 DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
+/* ITE 8893 has the same problem as the 8892 */
+DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8893, quirk_use_pcie_bridge_dma_alias);
 /* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
 DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
 
-- 
2.12.1

[PATCH] pci/quirks: ITE 8893 needs quirk_use_pcie_bridge_dma_alias

[Jarod Wilson]

This bridge has the same problems as the ITE 8892, which were resulting in
crippling an older PCI 1Gbps NIC down to 45Mbps throughput with IOMMU and
VT-d enabled. With the patch, this old e1000 goes back up to ~900Mbps.

Suggested-by: Alex Williamson 
CC: linux-...@vger.kernel.org
CC: Alex Williamson 
CC: Bjorn Helgaas 
Signed-off-by: Jarod Wilson 
---
 drivers/pci/quirks.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 9236e40ac055..c95fbf2431f8 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -3914,6 +3914,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ASMEDIA, 0x1080,
 DECLARE_PCI_FIXUP_HEADER(0x10e3, 0x8113, quirk_use_pcie_bridge_dma_alias);
 /* ITE 8892, https://bugzilla.kernel.org/show_bug.cgi?id=73551 */
 DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8892, quirk_use_pcie_bridge_dma_alias);
+/* ITE 8893 has the same problem as the 8892 */
+DECLARE_PCI_FIXUP_HEADER(0x1283, 0x8893, quirk_use_pcie_bridge_dma_alias);
 /* Intel 82801, https://bugzilla.kernel.org/show_bug.cgi?id=44881#c49 */
 DECLARE_PCI_FIXUP_HEADER(0x8086, 0x244e, quirk_use_pcie_bridge_dma_alias);
 
-- 
2.12.1