Skip to contentSkip to conversation

PCI: imx6: remove outbound io/mem ATU region mapping

RFC PATCH New Tim Harvey 6 participants 18 messages Oct 23, 2013 – Dec 5, 2013 PCI

lore
Share & tools
lore

Tim Harvey

author
#0 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

---
 drivers/pci/host/pcie-designware.c |   41 +++---------------------------------
 1 file changed, 3 insertions(+), 38 deletions(-)
drivers/pci/host/pcie-designware.c +3 -38
index 5d1f268..6cce779 100644
@@ -46,7 +46,6 @@
 #define PCIE_ATU_VIEWPORT		0x900
 #define PCIE_ATU_REGION_INBOUND		(0x1 << 31)
 #define PCIE_ATU_REGION_OUTBOUND	(0x0 << 31)
-#define PCIE_ATU_REGION_INDEX1		(0x1 << 0)
 #define PCIE_ATU_REGION_INDEX0		(0x0 << 0)
 #define PCIE_ATU_CR1			0x904
 #define PCIE_ATU_TYPE_MEM		(0x0 << 0)
@@ -502,8 +501,8 @@ static void dw_pcie_prog_viewport_cfg0(struct pcie_port *pp, u32 busdev)
 
 static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev)
 {
-	/* Program viewport 1 : OUTBOUND : CFG1 */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
+	/* Program viewport 0 : OUTBOUND : CFG1 */
+	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
 			  PCIE_ATU_VIEWPORT);
 	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
 	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
@@ -513,38 +512,8 @@ static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev)
 			  PCIE_ATU_LIMIT);
 	dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET);
 	dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET);
-}
-
-static void dw_pcie_prog_viewport_mem_outbound(struct pcie_port *pp)
-{
-	/* Program viewport 0 : OUTBOUND : MEM */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
-			  PCIE_ATU_VIEWPORT);
-	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_MEM, PCIE_ATU_CR1);
-	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
-	dw_pcie_writel_rc(pp, pp->mem_base, PCIE_ATU_LOWER_BASE);
-	dw_pcie_writel_rc(pp, (pp->mem_base >> 32), PCIE_ATU_UPPER_BASE);
-	dw_pcie_writel_rc(pp, pp->mem_base + pp->config.mem_size - 1,
-			  PCIE_ATU_LIMIT);
-	dw_pcie_writel_rc(pp, pp->config.mem_bus_addr, PCIE_ATU_LOWER_TARGET);
-	dw_pcie_writel_rc(pp, upper_32_bits(pp->config.mem_bus_addr),
-			  PCIE_ATU_UPPER_TARGET);
-}
-
-static void dw_pcie_prog_viewport_io_outbound(struct pcie_port *pp)
-{
-	/* Program viewport 1 : OUTBOUND : IO */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
-			  PCIE_ATU_VIEWPORT);
-	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_IO, PCIE_ATU_CR1);
+	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
 	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
-	dw_pcie_writel_rc(pp, pp->io_base, PCIE_ATU_LOWER_BASE);
-	dw_pcie_writel_rc(pp, (pp->io_base >> 32), PCIE_ATU_UPPER_BASE);
-	dw_pcie_writel_rc(pp, pp->io_base + pp->config.io_size - 1,
-			  PCIE_ATU_LIMIT);
-	dw_pcie_writel_rc(pp, pp->config.io_bus_addr, PCIE_ATU_LOWER_TARGET);
-	dw_pcie_writel_rc(pp, upper_32_bits(pp->config.io_bus_addr),
-			  PCIE_ATU_UPPER_TARGET);
 }
 
 static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
@@ -560,11 +529,9 @@ static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 	if (bus->parent->number == pp->root_bus_nr) {
 		dw_pcie_prog_viewport_cfg0(pp, busdev);
 		ret = cfg_read(pp->va_cfg0_base + address, where, size, val);
-		dw_pcie_prog_viewport_mem_outbound(pp);
 	} else {
 		dw_pcie_prog_viewport_cfg1(pp, busdev);
 		ret = cfg_read(pp->va_cfg1_base + address, where, size, val);
-		dw_pcie_prog_viewport_io_outbound(pp);
 	}
 
 	return ret;
@@ -583,11 +550,9 @@ static int dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 	if (bus->parent->number == pp->root_bus_nr) {
 		dw_pcie_prog_viewport_cfg0(pp, busdev);
 		ret = cfg_write(pp->va_cfg0_base + address, where, size, val);
-		dw_pcie_prog_viewport_mem_outbound(pp);
 	} else {
 		dw_pcie_prog_viewport_cfg1(pp, busdev);
 		ret = cfg_write(pp->va_cfg1_base + address, where, size, val);
-		dw_pcie_prog_viewport_io_outbound(pp);
 	}
 
 	return ret;
Signature · 1 line
1.7.9.5
#1 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Dear Tim Harvey,

Quoted 8 lines

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

Quoted 5 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

This probably needs further fixes to the DT on exynos and omap. And it also needs testing on those platforms as this was developed on freescale mx6.

Best regards,
Marek Vasut

#2 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

In reply to Tim Harvey

Hi,

Quoted 11 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

I would suggest to wait for Kishon's work who would be trying to optimize viewport allocations.

Regards
Pratyush

Quoted 95 lines
---
 drivers/pci/host/pcie-designware.c |   41 +++---------------------------------
 1 file changed, 3 insertions(+), 38 deletions(-)
drivers/pci/host/pcie-designware.c +3 -38
index 5d1f268..6cce779 100644
@@ -46,7 +46,6 @@
 #define PCIE_ATU_VIEWPORT		0x900
 #define PCIE_ATU_REGION_INBOUND		(0x1 << 31)
 #define PCIE_ATU_REGION_OUTBOUND	(0x0 << 31)
-#define PCIE_ATU_REGION_INDEX1		(0x1 << 0)
 #define PCIE_ATU_REGION_INDEX0		(0x0 << 0)
 #define PCIE_ATU_CR1			0x904
 #define PCIE_ATU_TYPE_MEM		(0x0 << 0)
@@ -502,8 +501,8 @@ static void dw_pcie_prog_viewport_cfg0(struct pcie_port *pp, u32 busdev)
 
 static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev)
 {
-	/* Program viewport 1 : OUTBOUND : CFG1 */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
+	/* Program viewport 0 : OUTBOUND : CFG1 */
+	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
 			  PCIE_ATU_VIEWPORT);
 	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
 	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
@@ -513,38 +512,8 @@ static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev)
 			  PCIE_ATU_LIMIT);
 	dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET);
 	dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET);
-}
-
-static void dw_pcie_prog_viewport_mem_outbound(struct pcie_port *pp)
-{
-	/* Program viewport 0 : OUTBOUND : MEM */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
-			  PCIE_ATU_VIEWPORT);
-	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_MEM, PCIE_ATU_CR1);
-	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
-	dw_pcie_writel_rc(pp, pp->mem_base, PCIE_ATU_LOWER_BASE);
-	dw_pcie_writel_rc(pp, (pp->mem_base >> 32), PCIE_ATU_UPPER_BASE);
-	dw_pcie_writel_rc(pp, pp->mem_base + pp->config.mem_size - 1,
-			  PCIE_ATU_LIMIT);
-	dw_pcie_writel_rc(pp, pp->config.mem_bus_addr, PCIE_ATU_LOWER_TARGET);
-	dw_pcie_writel_rc(pp, upper_32_bits(pp->config.mem_bus_addr),
-			  PCIE_ATU_UPPER_TARGET);
-}
-
-static void dw_pcie_prog_viewport_io_outbound(struct pcie_port *pp)
-{
-	/* Program viewport 1 : OUTBOUND : IO */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
-			  PCIE_ATU_VIEWPORT);
-	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_IO, PCIE_ATU_CR1);
+	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
 	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
-	dw_pcie_writel_rc(pp, pp->io_base, PCIE_ATU_LOWER_BASE);
-	dw_pcie_writel_rc(pp, (pp->io_base >> 32), PCIE_ATU_UPPER_BASE);
-	dw_pcie_writel_rc(pp, pp->io_base + pp->config.io_size - 1,
-			  PCIE_ATU_LIMIT);
-	dw_pcie_writel_rc(pp, pp->config.io_bus_addr, PCIE_ATU_LOWER_TARGET);
-	dw_pcie_writel_rc(pp, upper_32_bits(pp->config.io_bus_addr),
-			  PCIE_ATU_UPPER_TARGET);
 }
 
 static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
@@ -560,11 +529,9 @@ static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 	if (bus->parent->number == pp->root_bus_nr) {
 		dw_pcie_prog_viewport_cfg0(pp, busdev);
 		ret = cfg_read(pp->va_cfg0_base + address, where, size, val);
-		dw_pcie_prog_viewport_mem_outbound(pp);
 	} else {
 		dw_pcie_prog_viewport_cfg1(pp, busdev);
 		ret = cfg_read(pp->va_cfg1_base + address, where, size, val);
-		dw_pcie_prog_viewport_io_outbound(pp);
 	}
 
 	return ret;
@@ -583,11 +550,9 @@ static int dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 	if (bus->parent->number == pp->root_bus_nr) {
 		dw_pcie_prog_viewport_cfg0(pp, busdev);
 		ret = cfg_write(pp->va_cfg0_base + address, where, size, val);
-		dw_pcie_prog_viewport_mem_outbound(pp);
 	} else {
 		dw_pcie_prog_viewport_cfg1(pp, busdev);
 		ret = cfg_write(pp->va_cfg1_base + address, where, size, val);
-		dw_pcie_prog_viewport_io_outbound(pp);
 	}
 
 	return ret;
Signature · 1 line
1.7.9.5
#3 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Dear Pratyush Anand,

Quoted 11 lines

Hi,

Quoted 8 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

Quoted 7 lines
Quoted 3 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Best regards,
Marek Vasut

#4 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Quoted 17 lines · Marek Vasut

Dear Pratyush Anand,

Quoted 15 lines

Hi,

Quoted 11 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Quoted 5 lines
Quoted 3 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Regards
Pratyush

Quoted 3 lines

Best regards,
Marek Vasut

#5 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Hi Pratyush:

Quoted 28 lines

-----Original Message-----
From: Pratyush Anand [mailto:pratyush.anand at st.com]
Sent: Wednesday, November 27, 2013 11:51 AM
To: Marek Vasut
Cc: Tim Harvey; Kishon Vijay Abraham I; linux-pci at vger.kernel.org; linux-arm- kernel at lists.infradead.org; Bjorn Helgaas; Frank Li; Jingoo Han; Mohit KUMAR DCG; Zhu Richard-R65037; Sascha Hauer; Sean Cross; Shawn Guo; Siva Reddy Kallam; Srikanth T Shivanand; Troy Kisky; Yinghai Lu
Subject: Re: [PATCH RFC] PCI: imx6: remove outbound io/mem ATU region mapping

Quoted 17 lines · Marek Vasut

Dear Pratyush Anand,

Quoted 15 lines

Hi,

Quoted 11 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

[Richard] Regarding to the commit message, and tests at my side,
the switch wouldn't work well on imx6 platforms, if "the calls that inappropriately re-configures
the ATU viewport for outbound memory and IO after config cycles " are not removed.

Quoted 6 lines

Regards
Pratyush

Quoted 3 lines

Best regards,
Marek Vasut

Best Regards
Richard Zhu

#6 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

In reply to Pratyush Anand

Dear Pratyush Anand,

Quoted 19 lines · Marek Vasut
Quoted 18 lines

Dear Pratyush Anand,

Quoted 16 lines

Hi,

Quoted 12 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

Best regards,
Marek Vasut

#7 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Quoted 25 lines · Marek Vasut

Dear Pratyush Anand,

Quoted 23 lines · Marek Vasut
Quoted 22 lines

Dear Pratyush Anand,

Quoted 19 lines

Hi,

Quoted 5 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

Quoted 7 lines

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

Quoted 7 lines
Quoted 3 lines
Quoted 1 line

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

Regards
Pratyush

Quoted 3 lines

Best regards,
Marek Vasut

#8 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Hi Pratyush,

Quoted 31 lines · Marek Vasut
Quoted 30 lines

Dear Pratyush Anand,

Quoted 27 lines · Marek Vasut
Quoted 18 lines

Dear Pratyush Anand,

Quoted 16 lines

Hi,

Quoted 12 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Quoted 5 lines
Quoted 3 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Quoted 8 lines
Quoted 3 lines

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

Thanks!

Best regards,
Marek Vasut

#9 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

In reply to Pratyush Anand
Quoted 28 lines · Pratyush Anand
Quoted 27 lines · Marek Vasut
Quoted 26 lines · Marek Vasut
Quoted 18 lines

Dear Pratyush Anand,

Quoted 16 lines

Hi,

Quoted 12 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Quoted 5 lines
Quoted 3 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Quoted 8 lines
Quoted 3 lines

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

I agree with Pratyush Anand's opinion.
One of our engineers had tested PCI cards under switch. As far as I know, there was no issue about this. However, currently, I don't have any switches. So, I Cannot test this on Exynos platform.

Mohit,
If you test a switch on SPEAr platform, it will be very helpful. Thank you.

Best regards,
Jingoo Han

#10 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Dear Jingoo Han,

Quoted 39 lines · Pratyush Anand
Quoted 29 lines · Marek Vasut
Quoted 28 lines
Quoted 27 lines · Marek Vasut
Quoted 18 lines

Dear Pratyush Anand,

Quoted 16 lines

Hi,

Quoted 12 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Quoted 5 lines
Quoted 3 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Quoted 8 lines
Quoted 3 lines

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

Quoted 5 lines

I agree with Pratyush Anand's opinion.
One of our engineers had tested PCI cards under switch. As far as I know, there was no issue about this. However, currently, I don't have any switches. So, I Cannot test this on Exynos platform.

OK, I will wait for Richard to come up with confirmation if possible. Looks like he's out of the office, so it might take a bit.

Quoted 6 lines

Mohit,
If you test a switch on SPEAr platform, it will be very helpful. Thank you.

Best regards,
Jingoo Han

Best regards,
Marek Vasut

#11 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Hi Marek:

Quoted 53 lines

-----Original Message-----
From: Marek Vasut [mailto:marex at denx.de]
Sent: Friday, November 29, 2013 10:21 AM
To: Jingoo Han
Cc: 'Pratyush Anand'; 'Mohit KUMAR DCG'; 'Tim Harvey'; 'Kishon Vijay Abraham I'; linux-pci at vger.kernel.org; linux-arm-kernel at lists.infradead.org; 'Bjorn Helgaas'; 'Frank Li'; Zhu Richard-R65037; 'Sascha Hauer'; 'Sean Cross'; 'Shawn Guo'; 'Siva Reddy Kallam'; 'Srikanth T Shivanand'; 'Troy Kisky'; 'Yinghai Lu'
Subject: Re: [PATCH RFC] PCI: imx6: remove outbound io/mem ATU region mapping

Dear Jingoo Han,

Quoted 14 lines · Pratyush Anand
Quoted 13 lines · Marek Vasut
Quoted 12 lines
Quoted 12 lines · Marek Vasut
Quoted 11 lines

Dear Pratyush Anand,

Quoted 9 lines

Hi,

Quoted 6 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for
type0 and type1 config cycles but is not necessary for outbound io/mem regions.

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not

necessary.

Quoted 23 lines
Quoted 22 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
Quoted 4 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

Quoted 8 lines
Quoted 4 lines
Quoted 1 line

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

Quoted 9 lines
Quoted 5 lines

I agree with Pratyush Anand's opinion.
One of our engineers had tested PCI cards under switch.
As far as I know, there was no issue about this.
However, currently, I don't have any switches. So, I Cannot test this on Exynos platform.

OK, I will wait for Richard to come up with confirmation if possible. Looks like he's out of the office, so it might take a bit.

[Richard] One Pericom PI7C9X2G303EL pcie switch, and two pcie ep deivces(one is intel e1000e nic,
the other is one xhci device) are tested on imx6q sabresd board.
Without removing outbound io/mem regions view map during the cfg0/1 read/write cycle, both of these devices can't work well at my side.
Works well after remove them during the cfg0/1 read/write cycles.

Quoted 9 lines
Quoted 6 lines

Mohit,
If you test a switch on SPEAr platform, it will be very helpful. Thank you.

Best regards,
Jingoo Han

Best regards,
Marek Vasut

Best Regards
Richard Zhu

#12 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Dear Richard Zhu,

Quoted 63 lines

Hi Marek:

Quoted 62 lines

-----Original Message-----
From: Marek Vasut [mailto:marex at denx.de]
Sent: Friday, November 29, 2013 10:21 AM
To: Jingoo Han
Cc: 'Pratyush Anand'; 'Mohit KUMAR DCG'; 'Tim Harvey'; 'Kishon Vijay Abraham I'; linux-pci at vger.kernel.org;
linux-arm-kernel at lists.infradead.org; 'Bjorn Helgaas'; 'Frank Li'; Zhu Richard-R65037; 'Sascha Hauer'; 'Sean Cross'; 'Shawn Guo'; 'Siva Reddy Kallam'; 'Srikanth T Shivanand'; 'Troy Kisky'; 'Yinghai Lu' Subject: Re:
[PATCH RFC] PCI: imx6: remove outbound io/mem ATU region mapping

Dear Jingoo Han,

Quoted 14 lines · Pratyush Anand
Quoted 13 lines · Marek Vasut
Quoted 12 lines
Quoted 12 lines · Marek Vasut
Quoted 11 lines

Dear Pratyush Anand,

Quoted 9 lines

Hi,

Quoted 6 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for
type0 and type1 config cycles but is not necessary for outbound io/mem regions.

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not

necessary.

Quoted 27 lines
Quoted 13 lines
Quoted 13 lines
Quoted 12 lines
Quoted 9 lines
Quoted 7 lines
Quoted 3 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Quoted 8 lines
Quoted 3 lines

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

I agree with Pratyush Anand's opinion.
One of our engineers had tested PCI cards under switch.
As far as I know, there was no issue about this.
However, currently, I don't have any switches. So, I Cannot test this on Exynos platform.

Quoted 6 lines
Quoted 2 lines

OK, I will wait for Richard to come up with confirmation if possible. Looks like he's out of the office, so it might take a bit.

[Richard] One Pericom PI7C9X2G303EL pcie switch, and two pcie ep deivces(one is intel e1000e nic, the other is one xhci device) are tested on imx6q sabresd board.

How/what does have such pericom switch and can be attached to an MX6 sabresdp ? Where can I get it?

Quoted 3 lines

Without removing outbound io/mem regions view map during the cfg0/1 read/write cycle, both of these devices can't work well at my side. Works well after remove them during the cfg0/1 read/write cycles.

Understood. Given that the iATU programming works on other CPUs (confirmed on st spear and ti dra7xx), we might have some issues with the iATU on MX6 . Is there anything special about the iATU on the MX6 ?

#13 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Hi Marek:

Quoted 82 lines

-----Original Message-----
From: linux-pci-owner at vger.kernel.org [mailto:linux-pci-owner at vger.kernel.org] On Behalf Of Marek Vasut
Sent: Tuesday, December 03, 2013 5:20 PM
To: Zhu Richard-R65037
Cc: Jingoo Han; 'Pratyush Anand'; 'Mohit KUMAR DCG'; 'Tim Harvey'; 'Kishon Vijay Abraham I'; linux-pci at vger.kernel.org; linux-arm-
kernel at lists.infradead.org; 'Bjorn Helgaas'; 'Frank Li'; 'Sascha Hauer'; 'Sean Cross'; 'Shawn Guo'; 'Siva Reddy Kallam'; 'Srikanth T Shivanand'; 'Troy Kisky'; 'Yinghai Lu'
Subject: Re: [PATCH RFC] PCI: imx6: remove outbound io/mem ATU region mapping

Dear Richard Zhu,

Quoted 10 lines

Hi Marek:

Quoted 9 lines

-----Original Message-----
From: Marek Vasut [mailto:marex at denx.de]
Sent: Friday, November 29, 2013 10:21 AM
To: Jingoo Han
Cc: 'Pratyush Anand'; 'Mohit KUMAR DCG'; 'Tim Harvey'; 'Kishon Vijay Abraham I'; linux-pci at vger.kernel.org;
linux-arm-kernel at lists.infradead.org; 'Bjorn Helgaas'; 'Frank Li'; Zhu Richard-R65037; 'Sascha Hauer'; 'Sean Cross'; 'Shawn Guo'; 'Siva Reddy Kallam'; 'Srikanth T Shivanand'; 'Troy Kisky'; 'Yinghai Lu' Subject:

Re:

Quoted 23 lines
Quoted 23 lines

[PATCH RFC] PCI: imx6: remove outbound io/mem ATU region mapping

Dear Jingoo Han,

Quoted 18 lines · Pratyush Anand
Quoted 17 lines · Marek Vasut
Quoted 16 lines
Quoted 16 lines · Marek Vasut
Quoted 15 lines

Dear Pratyush Anand,

Quoted 13 lines

Hi,

Quoted 10 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for
type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not

Quoted 5 lines
Quoted 5 lines

necessary.

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other

configurations.

Quoted 28 lines
Quoted 27 lines
Quoted 19 lines
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Quoted 6 lines
Quoted 4 lines

It seems to me that in your controller you have only one view port. Also mem_base and mem_bus_addr is same. Thats why it works in your case.

MX6 has 4 In/4 Out viewports AFAICT.

Then if I do not miss anything, MX6 should work even without this patch.

Quoted 8 lines
Quoted 3 lines

Yes, yet, it does not. On the other hand, this defect is only problematic if you use the devices behind a switch. Do you use a PCIe switch on your platform please?

I had tested with device under bridge.
Mohit is in process of validating SPEAr patches with latest kernel. Mohit, see if we can arrange a switch and test, once SPEAr platform is ready with latest kernel.

Quoted 5 lines

I agree with Pratyush Anand's opinion.
One of our engineers had tested PCI cards under switch. As far as I know, there was no issue about this. However, currently, I don't have any switches. So, I Cannot test this on Exynos platform.

OK, I will wait for Richard to come up with confirmation if possible. Looks like he's out of the office, so it might take a bit.

Quoted 6 lines
Quoted 3 lines

[Richard] One Pericom PI7C9X2G303EL pcie switch, and two pcie ep deivces(one is intel e1000e nic, the other is one xhci device) are tested on imx6q sabresd board.

How/what does have such pericom switch and can be attached to an MX6 sabresdp ? Where can I get it?

[Richard] You can apply the example from Pericom.

Quoted 8 lines
Quoted 4 lines

Without removing outbound io/mem regions view map during the cfg0/1 read/write cycle, both of these devices can't work well at my side. Works well after remove them during the cfg0/1 read/write cycles.

Understood. Given that the iATU programming works on other CPUs (confirmed on st spear and ti dra7xx), we might have some issues with the iATU on MX6 . Is there anything special about the iATU on the MX6 ?

[Richard] As I know that there is no anything special about the iATU on MX6. Let me to make a double check with IC team later.

Quoted 4 lines

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo at vger.kernel.org More majordomo info at
http://vger.kernel.org/majordomo-info.html

Best Regards
Richard Zhu

#14 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

On Wednesday, December 04, 2013 at 03:38:49 AM, Richard Zhu wrote:
[...]

Quoted 8 lines
Quoted 6 lines
Quoted 3 lines

[Richard] One Pericom PI7C9X2G303EL pcie switch, and two pcie ep deivces(one is intel e1000e nic, the other is one xhci device) are tested on imx6q sabresd board.

How/what does have such pericom switch and can be attached to an MX6 sabresdp ? Where can I get it?

[Richard] You can apply the example from Pericom.

Which one? Please point me to a website or something here.

Quoted 10 lines
Quoted 7 lines
Quoted 3 lines

Without removing outbound io/mem regions view map during the cfg0/1 read/write cycle, both of these devices can't work well at my side. Works well after remove them during the cfg0/1 read/write cycles.

Understood. Given that the iATU programming works on other CPUs (confirmed on st spear and ti dra7xx), we might have some issues with the iATU on MX6 . Is there anything special about the iATU on the MX6 ?

[Richard] As I know that there is no anything special about the iATU on MX6. Let me to make a double check with IC team later.

Oh this would be absolutelly _amazing_ if you could do that. Thank you very much!
[...]
Best regards,
Marek Vasut

#15 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Hi Marek:

Quoted 17 lines

-----Original Message-----
From: Marek Vasut [mailto:marex at denx.de]
Sent: Wednesday, December 04, 2013 11:45 PM
To: Zhu Richard-R65037
Cc: Jingoo Han; 'Pratyush Anand'; 'Mohit KUMAR DCG'; 'Tim Harvey'; 'Kishon Vijay Abraham I'; linux-pci at vger.kernel.org; linux-arm-
kernel at lists.infradead.org; 'Bjorn Helgaas'; 'Frank Li'; 'Sascha Hauer'; 'Sean Cross'; 'Shawn Guo'; 'Siva Reddy Kallam'; 'Srikanth T Shivanand'; 'Troy Kisky'; 'Yinghai Lu'
Subject: Re: [PATCH RFC] PCI: imx6: remove outbound io/mem ATU region mapping

On Wednesday, December 04, 2013 at 03:38:49 AM, Richard Zhu wrote:
[...]

Quoted 4 lines
Quoted 4 lines
Quoted 3 lines

[Richard] One Pericom PI7C9X2G303EL pcie switch, and two pcie ep deivces(one is intel e1000e nic, the other is one xhci device) are tested on imx6q sabresd board.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

How/what does have such pericom switch and can be attached to an MX6 sabresdp ? Where can I get it?

[Richard] You can apply the example from Pericom.

Which one? Please point me to a website or something here.

[Richard] The model of the pcie switch used by me is PI7C9X2G303EL Ver3.0 provided by Pericom. Here is the url of this switch.
http://www.pericom.com/products/pcie-switch/?part=PI7C9X2G303EL

Quoted 10 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Without removing outbound io/mem regions view map during the cfg0/1 read/write cycle, both of these devices can't work well at my

side.

Quoted 5 lines
Quoted 5 lines
Quoted 1 line

Works well after remove them during the cfg0/1 read/write cycles.

Understood. Given that the iATU programming works on other CPUs (confirmed on st spear and ti dra7xx), we might have some issues with the iATU on MX6 . Is there anything special about the iATU on the

MX6 ?

Quoted 1 line
Quoted 8 lines
Quoted 2 lines

[Richard] As I know that there is no anything special about the iATU on MX6. Let me to make a double check with IC team later.

Oh this would be absolutelly _amazing_ if you could do that. Thank you very much!
[...]
Best regards,
Marek Vasut

Best Regards
Richard Zhu

maintainer
#16 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

In reply to Tim Harvey
Quoted 8 lines · Tim Harvey

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

Quoted 5 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Dropping for now, per Marek and Pratyush's concerns. Poke me if/when those are resolved.

Quoted 94 lines
---
 drivers/pci/host/pcie-designware.c |   41 +++---------------------------------
 1 file changed, 3 insertions(+), 38 deletions(-)
drivers/pci/host/pcie-designware.c +3 -38
index 5d1f268..6cce779 100644
@@ -46,7 +46,6 @@
 #define PCIE_ATU_VIEWPORT		0x900
 #define PCIE_ATU_REGION_INBOUND		(0x1 << 31)
 #define PCIE_ATU_REGION_OUTBOUND	(0x0 << 31)
-#define PCIE_ATU_REGION_INDEX1		(0x1 << 0)
 #define PCIE_ATU_REGION_INDEX0		(0x0 << 0)
 #define PCIE_ATU_CR1			0x904
 #define PCIE_ATU_TYPE_MEM		(0x0 << 0)
@@ -502,8 +501,8 @@ static void dw_pcie_prog_viewport_cfg0(struct pcie_port *pp, u32 busdev)
 
 static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev)
 {
-	/* Program viewport 1 : OUTBOUND : CFG1 */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
+	/* Program viewport 0 : OUTBOUND : CFG1 */
+	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
 			  PCIE_ATU_VIEWPORT);
 	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
 	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
@@ -513,38 +512,8 @@ static void dw_pcie_prog_viewport_cfg1(struct pcie_port *pp, u32 busdev)
 			  PCIE_ATU_LIMIT);
 	dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET);
 	dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET);
-}
-
-static void dw_pcie_prog_viewport_mem_outbound(struct pcie_port *pp)
-{
-	/* Program viewport 0 : OUTBOUND : MEM */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
-			  PCIE_ATU_VIEWPORT);
-	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_MEM, PCIE_ATU_CR1);
-	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
-	dw_pcie_writel_rc(pp, pp->mem_base, PCIE_ATU_LOWER_BASE);
-	dw_pcie_writel_rc(pp, (pp->mem_base >> 32), PCIE_ATU_UPPER_BASE);
-	dw_pcie_writel_rc(pp, pp->mem_base + pp->config.mem_size - 1,
-			  PCIE_ATU_LIMIT);
-	dw_pcie_writel_rc(pp, pp->config.mem_bus_addr, PCIE_ATU_LOWER_TARGET);
-	dw_pcie_writel_rc(pp, upper_32_bits(pp->config.mem_bus_addr),
-			  PCIE_ATU_UPPER_TARGET);
-}
-
-static void dw_pcie_prog_viewport_io_outbound(struct pcie_port *pp)
-{
-	/* Program viewport 1 : OUTBOUND : IO */
-	dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
-			  PCIE_ATU_VIEWPORT);
-	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_IO, PCIE_ATU_CR1);
+	dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
 	dw_pcie_writel_rc(pp, PCIE_ATU_ENABLE, PCIE_ATU_CR2);
-	dw_pcie_writel_rc(pp, pp->io_base, PCIE_ATU_LOWER_BASE);
-	dw_pcie_writel_rc(pp, (pp->io_base >> 32), PCIE_ATU_UPPER_BASE);
-	dw_pcie_writel_rc(pp, pp->io_base + pp->config.io_size - 1,
-			  PCIE_ATU_LIMIT);
-	dw_pcie_writel_rc(pp, pp->config.io_bus_addr, PCIE_ATU_LOWER_TARGET);
-	dw_pcie_writel_rc(pp, upper_32_bits(pp->config.io_bus_addr),
-			  PCIE_ATU_UPPER_TARGET);
 }
 
 static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
@@ -560,11 +529,9 @@ static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 	if (bus->parent->number == pp->root_bus_nr) {
 		dw_pcie_prog_viewport_cfg0(pp, busdev);
 		ret = cfg_read(pp->va_cfg0_base + address, where, size, val);
-		dw_pcie_prog_viewport_mem_outbound(pp);
 	} else {
 		dw_pcie_prog_viewport_cfg1(pp, busdev);
 		ret = cfg_read(pp->va_cfg1_base + address, where, size, val);
-		dw_pcie_prog_viewport_io_outbound(pp);
 	}
 
 	return ret;
@@ -583,11 +550,9 @@ static int dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 	if (bus->parent->number == pp->root_bus_nr) {
 		dw_pcie_prog_viewport_cfg0(pp, busdev);
 		ret = cfg_write(pp->va_cfg0_base + address, where, size, val);
-		dw_pcie_prog_viewport_mem_outbound(pp);
 	} else {
 		dw_pcie_prog_viewport_cfg1(pp, busdev);
 		ret = cfg_write(pp->va_cfg1_base + address, where, size, val);
-		dw_pcie_prog_viewport_io_outbound(pp);
 	}
 
 	return ret;
Signature · 1 line
1.7.9.5
#17 lore raw

RFC PATCH 1/1 PCI: imx6: remove outbound io/mem ATU region mapping

Dear Bjorn Helgaas,

Quoted 9 lines · Tim Harvey
Quoted 8 lines

The IMX6 iATU is used for address translation between the AXI bus address space and PCI address space. This is used for type0 and type1 config cycles but is not necessary for outbound io/mem regions.

This patch removes the calls that inappropriately re-configures the ATU viewport for outbound memory and IO after config cycles and removes them altogether as they are not necessary.

Quoted 8 lines
Quoted 5 lines

This resolves issues with PCI devices behind switches and has been tested with a Gige device behind a PLX PEX860x switch. More testing is needed for other configurations.

Dropping for now, per Marek and Pratyush's concerns. Poke me if/when those are resolved.

I hope (unless of course I again fail to under load of other stuff to do) to look at the PCIe stuff again next week once I'm in the office. In the meantime, I'd love to submit new set of patches.

Best regards,
Marek Vasut