Skip to contentSkip to conversation

Add PCIe root complex support for AM654x SoC

PATCH New Sekhar Nori 4 participants 21 messages Aug 1, 2019 – Aug 13, 2019

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

v3 cover Add PCIe root complex support for AM654x SoC

Hi,

This patch series adds PCIe root complex support for AM654x SoC.

The device-tree files are based on bindings accepted in linux. See files Documentation/devicetree/bindings/phy/ti,phy-am654-serdes.txt and Documentation/devicetree/bindings/pci/pci-keystone.txt in latest mainline master.

I have not posted the actual board-specific device-tree bits yet. The reason is that PCIe slot is on a daughter card on the AM65x EVM. I want to see how we can support that as an overlay in U-Boot. That needs some more attention. Meanwhile I have tested this using a patch that simply enables PCIe in the baseboard device-tree file itself.

v3:
- fix unit test for clk_is_match()

v2:
- add unit tests for new clock and dm APIs introduced
- add Lokesh's Reviewed-by: for patches #1 and #2

Sekhar Nori (7):
clk: add support for clk_is_match()
dm: core: add support for getting register address and size
pcie: ti: add driver for AM65x PCIe RC
phy: add support for AM654x SERDES
configs: am65x_evm_a53: enable PCIe support
arm: dts: k3-am65: add support for PCIe and SERDES
configs: am65x_evm_a53: enable support for PCIe ethernet cards

 arch/arm/dts/k3-am65-main.dtsi             | 108 +++++
 arch/arm/dts/k3-am65.dtsi                  |   1 +
 configs/am65x_evm_a53_defconfig            |   8 +
 drivers/clk/clk-uclass.c                   |  13 +
 drivers/core/fdtaddr.c                     |  17 +
 drivers/core/read.c                        |  20 +
 drivers/pci/Kconfig                        |   6 +
 drivers/pci/Makefile                       |   1 +
 drivers/pci/pcie_dw_ti.c                   | 725 +++++++++++++++++++++++++++++
 drivers/phy/Kconfig                        |   9 +
 drivers/phy/Makefile                       |   1 +
 drivers/phy/phy-ti-am654.c                 | 411 ++++++++++++++++
 include/clk.h                              |  12 +
 include/dm/fdtaddr.h                       |  18 +
 include/dm/read.h                          |  41 ++
 include/dt-bindings/phy/phy-am654-serdes.h |  13 +
 test/dm/clk.c                              |   1 +
 test/dm/test-fdt.c                         |  16 +-
 18 files changed, 1417 insertions(+), 4 deletions(-)
 create mode 100644 drivers/pci/pcie_dw_ti.c
 create mode 100644 drivers/phy/phy-ti-am654.c
 create mode 100644 include/dt-bindings/phy/phy-am654-serdes.h
Signature · 1 line
2.16.2
#1 lore raw

PATCH v3 1/7 clk: add support for clk_is_match()

Add support for clk_is_match() which is required to know if two clock pointers point to the same exact physical clock.

Also add a unit test for the new API.

---
 drivers/clk/clk-uclass.c | 13 +++++++++++++
 include/clk.h            | 12 ++++++++++++
 test/dm/clk.c            |  1 +
 3 files changed, 26 insertions(+)
drivers/clk/clk-uclass.c +13 -0
index 85dfe712f5ac..1409c03ea174 100644
@@ -515,6 +515,19 @@ int clk_get_by_id(ulong id, struct clk **clkp)
 	return -ENOENT;
 }
 
+bool clk_is_match(const struct clk *p, const struct clk *q)
+{
+	/* trivial case: identical struct clk's or both NULL */
+	if (p == q)
+		return true;
+
+	/* same device, id and data */
+	if (p->dev == q->dev && p->id == q->id && p->data == q->data)
+		return true;
+
+	return false;
+}
+
 UCLASS_DRIVER(clk) = {
 	.id		= UCLASS_CLK,
 	.name		= "clk",
include/clk.h +12 -0
index f8f56d9cf01d..f50e5caa2206 100644
@@ -333,6 +333,18 @@ int clk_disable(struct clk *clk);
  */
 int clk_disable_bulk(struct clk_bulk *bulk);
 
+/**
+ * clk_is_match - check if two clk's point to the same hardware clock
+ * @p: clk compared against q
+ * @q: clk compared against p
+ *
+ * Returns true if the two struct clk pointers both point to the same hardware
+ * clock node.
+ *
+ * Returns false otherwise. Note that two NULL clks are treated as matching.
+ */
+bool clk_is_match(const struct clk *p, const struct clk *q);
+
 int soc_clk_dump(void);
 
 /**
test/dm/clk.c +1 -0
index f301ecbb459d..676ef217f093 100644
@@ -24,6 +24,7 @@ static int dm_test_clk_base(struct unit_test_state *uts)
 	/* Get the same clk port in 2 different ways and compare */
 	ut_assertok(clk_get_by_index(dev, 1, &clk_method1));
 	ut_assertok(clk_get_by_index_nodev(dev_ofnode(dev), 1, &clk_method2));
+	ut_asserteq(clk_is_match(&clk_method1, &clk_method2), true);
 	ut_asserteq(clk_method1.id, clk_method2.id);
 
 	return 0;
Signature · 1 line
2.16.2
#2 lore raw

PATCH v3 1/7 clk: add support for clk_is_match()

Hi Sekhar,

Quoted 4 lines · Sekhar Nori

Add support for clk_is_match() which is required to know if two clock pointers point to the same exact physical clock.

I would suggest adding a note that this is inspired from the clk_is_match() implementation from the Linux kernel.

Thanks

#3 lore raw

PATCH v3 1/7 clk: add support for clk_is_match()

Hi Fabio,

Quoted 10 lines · Fabio Estevam

Hi Sekhar,

Quoted 4 lines · Sekhar Nori

Add support for clk_is_match() which is required to know if two clock pointers point to the same exact physical clock.

I would suggest adding a note that this is inspired from the clk_is_match() implementation from the Linux kernel.

Sure. I assume this goes into the commit description?

Tom, should I resend or this is something that you can fix while applying?

Thanks,
Sekhar

#4 lore raw

PATCH v3 1/7 clk: add support for clk_is_match()

Hi Sekhar,

Quoted 1 line · Sekhar Nori

Sure. I assume this goes into the commit description?

Yes, correct. Thanks

#5 lore raw

PATCH v3 1/7 clk: add support for clk_is_match()

In reply to Sekhar Nori
Quoted 11 lines · Sekhar Nori

Hi Fabio,

Quoted 8 lines · Fabio Estevam

Hi Sekhar,

Quoted 4 lines · Sekhar Nori

Add support for clk_is_match() which is required to know if two clock pointers point to the same exact physical clock.

Quoted 6 lines
Quoted 2 lines

I would suggest adding a note that this is inspired from the clk_is_match() implementation from the Linux kernel.

Sure. I assume this goes into the commit description?

Tom, should I resend or this is something that you can fix while applying?

Please v4 this part as I'm sure I'll otherwise forget, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190801/c115fd6e/attachment.sig>
#6 lore raw

PATCH v3 1/7 clk: add support for clk_is_match()

In reply to Sekhar Nori
Quoted 8 lines · Sekhar Nori

Add support for clk_is_match() which is required to know if two clock pointers point to the same exact physical clock.

Also add a unit test for the new API.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/914dadcb/attachment.sig>
#7 lore raw

PATCH v3 2/7 dm: core: add support for getting register address and size

In reply to Sekhar Nori

Current dev_read_*() API lacks support to get address and size of a "reg" property by name or index. Add support for the same.

Livetree support has been added but not tested on real hardware.

The existing unit tests testing reading address from device-tree have been updated to test address as well as size.

---
 drivers/core/fdtaddr.c | 17 +++++++++++++++++
 drivers/core/read.c    | 20 ++++++++++++++++++++
 include/dm/fdtaddr.h   | 18 ++++++++++++++++++
 include/dm/read.h      | 41 +++++++++++++++++++++++++++++++++++++++++
 test/dm/test-fdt.c     | 16 ++++++++++++----
 5 files changed, 108 insertions(+), 4 deletions(-)
drivers/core/fdtaddr.c +17 -0
index c2873861dacd..6850003a287b 100644
@@ -129,6 +129,23 @@ fdt_addr_t devfdt_get_addr_name(struct udevice *dev, const char *name)
 #endif
 }
 
+fdt_addr_t devfdt_get_addr_size_name(struct udevice *dev, const char *name,
+				     fdt_size_t *size)
+{
+#if CONFIG_IS_ENABLED(OF_CONTROL)
+	int index;
+
+	index = fdt_stringlist_search(gd->fdt_blob, dev_of_offset(dev),
+				      "reg-names", name);
+	if (index < 0)
+		return index;
+
+	return devfdt_get_addr_size_index(dev, index, size);
+#else
+	return FDT_ADDR_T_NONE;
+#endif
+}
+
 fdt_addr_t devfdt_get_addr(struct udevice *dev)
 {
 	return devfdt_get_addr_index(dev, 0);
drivers/core/read.c +20 -0
index 1a044b05e884..8b5502de1159 100644
@@ -82,6 +82,15 @@ fdt_addr_t dev_read_addr_index(struct udevice *dev, int index)
 		return devfdt_get_addr_index(dev, index);
 }
 
+fdt_addr_t dev_read_addr_size_index(struct udevice *dev, int index,
+				    fdt_size_t *size)
+{
+	if (ofnode_is_np(dev_ofnode(dev)))
+		return ofnode_get_addr_size_index(dev_ofnode(dev), index, size);
+	else
+		return devfdt_get_addr_size_index(dev, index, size);
+}
+
 void *dev_remap_addr_index(struct udevice *dev, int index)
 {
 	fdt_addr_t addr = dev_read_addr_index(dev, index);
@@ -102,6 +111,17 @@ fdt_addr_t dev_read_addr_name(struct udevice *dev, const char *name)
 		return dev_read_addr_index(dev, index);
 }
 
+fdt_addr_t dev_read_addr_size_name(struct udevice *dev, const char *name,
+				   fdt_size_t *size)
+{
+	int index = dev_read_stringlist_search(dev, "reg-names", name);
+
+	if (index < 0)
+		return FDT_ADDR_T_NONE;
+	else
+		return dev_read_addr_size_index(dev, index, size);
+}
+
 void *dev_remap_addr_name(struct udevice *dev, const char *name)
 {
 	fdt_addr_t addr = dev_read_addr_name(dev, name);
include/dm/fdtaddr.h +18 -0
index 3bc2599b6cbd..57b326cb3362 100644
@@ -120,4 +120,22 @@ fdt_addr_t devfdt_get_addr_size_index(struct udevice *dev, int index,
  */
 fdt_addr_t devfdt_get_addr_name(struct udevice *dev, const char *name);
 
+/**
+ * devfdt_get_addr_size_name() - Get the reg property and its size for a device,
+ *				 indexed by name
+ *
+ * Returns the address and size specified in the 'reg' property of a device.
+ *
+ * @dev: Pointer to a device
+ * @name: the 'reg' property can hold a list of <addr, size> pairs, with the
+ *	  'reg-names' property providing named-based identification. @index
+ *	  indicates the value to search for in 'reg-names'.
+ * @size: Pointer to size variable - this function returns the size
+ *        specified in the 'reg' property here
+ *
+ * @return addr
+ */
+fdt_addr_t devfdt_get_addr_size_name(struct udevice *dev, const char *name,
+				     fdt_size_t *size);
+
 #endif
include/dm/read.h +41 -0

65 lines not on this page · raw message

test/dm/test-fdt.c +12 -4
index ad8591639da5..1fb8b5c248d3 100644
@@ -549,12 +549,14 @@ static int dm_test_fdt_remap_addr_index_flat(struct unit_test_state *uts)
 {
 	struct udevice *dev;
 	fdt_addr_t addr;
+	fdt_size_t size;
 	void *paddr;
 
 	ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-	addr = devfdt_get_addr_index(dev, 0);
+	addr = devfdt_get_addr_size_index(dev, 0, &size);
 	ut_asserteq(0x8000, addr);
+	ut_asserteq(0x1000, size);
 
 	paddr = map_physmem(addr, 0, MAP_NOCACHE);
 	ut_assertnonnull(paddr);
@@ -569,12 +571,14 @@ static int dm_test_fdt_remap_addr_name_flat(struct unit_test_state *uts)
 {
 	struct udevice *dev;
 	fdt_addr_t addr;
+	fdt_size_t size;
 	void *paddr;
 
 	ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-	addr = devfdt_get_addr_name(dev, "sandbox-dummy-0");
+	addr = devfdt_get_addr_size_name(dev, "sandbox-dummy-0", &size);
 	ut_asserteq(0x8000, addr);
+	ut_asserteq(0x1000, size);
 
 	paddr = map_physmem(addr, 0, MAP_NOCACHE);
 	ut_assertnonnull(paddr);
@@ -609,12 +613,14 @@ static int dm_test_fdt_remap_addr_index_live(struct unit_test_state *uts)
 {
 	struct udevice *dev;
 	fdt_addr_t addr;
+	fdt_size_t size;
 	void *paddr;
 
 	ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-	addr = dev_read_addr_index(dev, 0);
+	addr = dev_read_addr_size_index(dev, 0, &size);
 	ut_asserteq(0x8000, addr);
+	ut_asserteq(0x1000, size);
 
 	paddr = map_physmem(addr, 0, MAP_NOCACHE);
 	ut_assertnonnull(paddr);
@@ -629,12 +635,14 @@ static int dm_test_fdt_remap_addr_name_live(struct unit_test_state *uts)
 {
 	struct udevice *dev;
 	fdt_addr_t addr;
+	fdt_size_t size;
 	void *paddr;
 
 	ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-	addr = dev_read_addr_name(dev, "sandbox-dummy-0");
+	addr = dev_read_addr_size_name(dev, "sandbox-dummy-0", &size);
 	ut_asserteq(0x8000, addr);
+	ut_asserteq(0x1000, size);
 
 	paddr = map_physmem(addr, 0, MAP_NOCACHE);
 	ut_assertnonnull(paddr);
Signature · 1 line
2.16.2
#8 lore raw

PATCH v3 2/7 dm: core: add support for getting register address and size

Quoted 68 lines · Sekhar Nori

Current dev_read_*() API lacks support to get address and size of a "reg" property by name or index. Add support for the same.

Livetree support has been added but not tested on real hardware.

The existing unit tests testing reading address from device-tree have been updated to test address as well as size.

---
 drivers/core/fdtaddr.c | 17 +++++++++++++++++
 drivers/core/read.c    | 20 ++++++++++++++++++++
 include/dm/fdtaddr.h   | 18 ++++++++++++++++++
 include/dm/read.h      | 41 +++++++++++++++++++++++++++++++++++++++++
 test/dm/test-fdt.c     | 16 ++++++++++++----
 5 files changed, 108 insertions(+), 4 deletions(-)
drivers/core/fdtaddr.c +17 -0
index c2873861dacd..6850003a287b 100644
@@ -129,6 +129,23 @@ fdt_addr_t devfdt_get_addr_name(struct udevice *dev, const char *name)
 #endif
 }
 
+fdt_addr_t devfdt_get_addr_size_name(struct udevice *dev, const char *name,
+                                    fdt_size_t *size)
+{
+#if CONFIG_IS_ENABLED(OF_CONTROL)
+       int index;
+
+       index = fdt_stringlist_search(gd->fdt_blob, dev_of_offset(dev),
+                                     "reg-names", name);
+       if (index < 0)
+               return index;
+
+       return devfdt_get_addr_size_index(dev, index, size);
+#else
+       return FDT_ADDR_T_NONE;
+#endif
+}
+
 fdt_addr_t devfdt_get_addr(struct udevice *dev)
 {
        return devfdt_get_addr_index(dev, 0);
drivers/core/read.c +9 -0
index 1a044b05e884..8b5502de1159 100644
@@ -82,6 +82,15 @@ fdt_addr_t dev_read_addr_index(struct udevice *dev, int index)
                return devfdt_get_addr_index(dev, index);
 }
 
+fdt_addr_t dev_read_addr_size_index(struct udevice *dev, int index,
+                                   fdt_size_t *size)
+{
+       if (ofnode_is_np(dev_ofnode(dev)))
+               return ofnode_get_addr_size_index(dev_ofnode(dev), index, size);
+       else
+               return devfdt_get_addr_size_index(dev, index, size);
+}
+
 void *dev_remap_addr_index(struct udevice *dev, int index)
 {
        fdt_addr_t addr = dev_read_addr_index(dev, index);

drivers/core/read.c fdt_addr_t dev_read_addr_name(struct udevice *dev, const char *name) · line 112

Quoted 2 lines
drivers/core/read.c +0 -0
@@ -102,6 +111,17 @@ fdt_addr_t dev_read_addr_name(struct udevice *dev, const char *name)
                return dev_read_addr_index(dev, index);
 }

maybe it makes sense to update the dev_remap_* variants to utilise the new dev_read_addr_* to be able to pass a valid size to map_physmem()? Currently a size of 0 is passed due to the lack of function which you are adding now.

Quoted 189 lines
drivers/core/read.c +11 -0
+fdt_addr_t dev_read_addr_size_name(struct udevice *dev, const char *name,
+                                  fdt_size_t *size)
+{
+       int index = dev_read_stringlist_search(dev, "reg-names", name);
+
+       if (index < 0)
+               return FDT_ADDR_T_NONE;
+       else
+               return dev_read_addr_size_index(dev, index, size);
+}
+
 void *dev_remap_addr_name(struct udevice *dev, const char *name)
 {
        fdt_addr_t addr = dev_read_addr_name(dev, name);
include/dm/fdtaddr.h +18 -0
index 3bc2599b6cbd..57b326cb3362 100644
@@ -120,4 +120,22 @@ fdt_addr_t devfdt_get_addr_size_index(struct udevice *dev, int index,
  */
 fdt_addr_t devfdt_get_addr_name(struct udevice *dev, const char *name);
 
+/**
+ * devfdt_get_addr_size_name() - Get the reg property and its size for a device,
+ *                              indexed by name
+ *
+ * Returns the address and size specified in the 'reg' property of a device.
+ *
+ * @dev: Pointer to a device
+ * @name: the 'reg' property can hold a list of <addr, size> pairs, with the
+ *       'reg-names' property providing named-based identification. @index
+ *       indicates the value to search for in 'reg-names'.
+ * @size: Pointer to size variable - this function returns the size
+ *        specified in the 'reg' property here
+ *
+ * @return addr
+ */
+fdt_addr_t devfdt_get_addr_size_name(struct udevice *dev, const char *name,
+                                    fdt_size_t *size);
+
 #endif
include/dm/read.h +41 -0
index 6ecd062e200c..0c62d62f1148 100644
@@ -144,6 +144,19 @@ int dev_read_size(struct udevice *dev, const char *propname);
  */
 fdt_addr_t dev_read_addr_index(struct udevice *dev, int index);
 
+/**
+ * dev_read_addr_size_index() - Get the indexed reg property of a device
+ *
+ * @dev: Device to read from
+ * @index: the 'reg' property can hold a list of <addr, size> pairs
+ *        and @index is used to select which one is required
+ * @size: place to put size value (on success)
+ *
+ * @return address or FDT_ADDR_T_NONE if not found
+ */
+fdt_addr_t dev_read_addr_size_index(struct udevice *dev, int index,
+                                   fdt_size_t *size);
+
 /**
  * dev_remap_addr_index() - Get the indexed reg property of a device
  *                               as a memory-mapped I/O pointer
@@ -168,6 +181,20 @@ void *dev_remap_addr_index(struct udevice *dev, int index);
  */
 fdt_addr_t dev_read_addr_name(struct udevice *dev, const char* name);
 
+/**
+ * dev_read_addr_size_name() - Get the reg property of a device, indexed by name
+ *
+ * @dev: Device to read from
+ * @name: the 'reg' property can hold a list of <addr, size> pairs, with the
+ *       'reg-names' property providing named-based identification. @index
+ *       indicates the value to search for in 'reg-names'.
+ *  @size: place to put size value (on success)
+ *
+ * @return address or FDT_ADDR_T_NONE if not found
+ */
+fdt_addr_t dev_read_addr_size_name(struct udevice *dev, const char *name,
+                                  fdt_size_t *size);
+
 /**
  * dev_remap_addr_name() - Get the reg property of a device, indexed by name,
  *                         as a memory-mapped I/O pointer
@@ -601,12 +628,26 @@ static inline fdt_addr_t dev_read_addr_index(struct udevice *dev, int index)
        return devfdt_get_addr_index(dev, index);
 }
 
+static inline fdt_addr_t dev_read_addr_size_index(struct udevice *dev,
+                                                 int index,
+                                                 fdt_size_t *size)
+{
+       return devfdt_get_addr_size_index(dev, index, size);
+}
+
 static inline fdt_addr_t dev_read_addr_name(struct udevice *dev,
                                            const char *name)
 {
        return devfdt_get_addr_name(dev, name);
 }
 
+static inline fdt_addr_t dev_read_addr_size_name(struct udevice *dev,
+                                                const char *name,
+                                                fdt_size_t *size)
+{
+       return devfdt_get_addr_size_name(dev, name, size);
+}
+
 static inline fdt_addr_t dev_read_addr(struct udevice *dev)
 {
        return devfdt_get_addr(dev);
test/dm/test-fdt.c +12 -5
index ad8591639da5..1fb8b5c248d3 100644
@@ -549,12 +549,14 @@ static int dm_test_fdt_remap_addr_index_flat(struct unit_test_state *uts)
 {
        struct udevice *dev;
        fdt_addr_t addr;
+       fdt_size_t size;
        void *paddr;
 
        ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-       addr = devfdt_get_addr_index(dev, 0);
+       addr = devfdt_get_addr_size_index(dev, 0, &size);
        ut_asserteq(0x8000, addr);
+       ut_asserteq(0x1000, size);
 
        paddr = map_physmem(addr, 0, MAP_NOCACHE);
        ut_assertnonnull(paddr);
@@ -569,12 +571,14 @@ static int dm_test_fdt_remap_addr_name_flat(struct unit_test_state *uts)
 {
        struct udevice *dev;
        fdt_addr_t addr;
+       fdt_size_t size;
        void *paddr;
 
        ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-       addr = devfdt_get_addr_name(dev, "sandbox-dummy-0");
+       addr = devfdt_get_addr_size_name(dev, "sandbox-dummy-0", &size);
        ut_asserteq(0x8000, addr);
+       ut_asserteq(0x1000, size);
 
        paddr = map_physmem(addr, 0, MAP_NOCACHE);
        ut_assertnonnull(paddr);
@@ -609,12 +613,14 @@ static int dm_test_fdt_remap_addr_index_live(struct unit_test_state *uts)
 {
        struct udevice *dev;
        fdt_addr_t addr;
+       fdt_size_t size;
        void *paddr;
 
        ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-       addr = dev_read_addr_index(dev, 0);
+       addr = dev_read_addr_size_index(dev, 0, &size);
        ut_asserteq(0x8000, addr);
+       ut_asserteq(0x1000, size);
 
        paddr = map_physmem(addr, 0, MAP_NOCACHE);
        ut_assertnonnull(paddr);
@@ -629,12 +635,14 @@ static int dm_test_fdt_remap_addr_name_live(struct unit_test_state *uts)
 {
        struct udevice *dev;
        fdt_addr_t addr;
+       fdt_size_t size;
        void *paddr;
 
        ut_assertok(uclass_find_device_by_seq(UCLASS_TEST_DUMMY, 0, true, &dev));
 
-       addr = dev_read_addr_name(dev, "sandbox-dummy-0");
+       addr = dev_read_addr_size_name(dev, "sandbox-dummy-0", &size);
        ut_asserteq(0x8000, addr);
+       ut_asserteq(0x1000, size);
 
        paddr = map_physmem(addr, 0, MAP_NOCACHE);
        ut_assertnonnull(paddr);
--

2.16.2

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

Signature · 1 line
- Daniel
#9 lore raw

PATCH v3 2/7 dm: core: add support for getting register address and size

Hi Daniel,

Quoted 20 lines · Daniel Schwierzeck
Quoted 20 lines
drivers/core/read.c +9 -0
index 1a044b05e884..8b5502de1159 100644
@@ -82,6 +82,15 @@ fdt_addr_t dev_read_addr_index(struct udevice *dev, int index)
                return devfdt_get_addr_index(dev, index);
 }
 
+fdt_addr_t dev_read_addr_size_index(struct udevice *dev, int index,
+                                   fdt_size_t *size)
+{
+       if (ofnode_is_np(dev_ofnode(dev)))
+               return ofnode_get_addr_size_index(dev_ofnode(dev), index, size);
+       else
+               return devfdt_get_addr_size_index(dev, index, size);
+}
+
 void *dev_remap_addr_index(struct udevice *dev, int index)
 {
        fdt_addr_t addr = dev_read_addr_index(dev, index);
Quoted 8 lines

drivers/core/read.c fdt_addr_t dev_read_addr_name(struct udevice *dev, const char *name) · line 112

Quoted 3 lines
drivers/core/read.c +0 -0
@@ -102,6 +111,17 @@ fdt_addr_t dev_read_addr_name(struct udevice *dev, const char *name)
                return dev_read_addr_index(dev, index);
 }

maybe it makes sense to update the dev_remap_* variants to utilise the new dev_read_addr_* to be able to pass a valid size to map_physmem()? Currently a size of 0 is passed due to the lack of function which you are adding now.

Thanks for pointing that out. I think this is something that can come as a follow-on series. This series itself was supposed to go into 2019.10, but held up due to a stupid bug I introduced.

I will send a v4 today with a commit message change that was asked with the hope that it can get into 2019.10 still.

Thanks,
Sekhar

#10 lore raw

PATCH v3 2/7 dm: core: add support for getting register address and size

In reply to Sekhar Nori
Quoted 10 lines · Sekhar Nori

Current dev_read_*() API lacks support to get address and size of a "reg" property by name or index. Add support for the same.

Livetree support has been added but not tested on real hardware.

The existing unit tests testing reading address from device-tree have been updated to test address as well as size.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/bd4f8a4f/attachment.sig>
#11 lore raw

PATCH v3 3/7 pcie: ti: add driver for AM65x PCIe RC

In reply to Sekhar Nori

Add driver supporting PCIe root-complex available
on TI's AM65x SoC.

---
 drivers/pci/Kconfig      |   6 +
 drivers/pci/Makefile     |   1 +
 drivers/pci/pcie_dw_ti.c | 725 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 732 insertions(+)
 create mode 100644 drivers/pci/pcie_dw_ti.c
drivers/pci/Kconfig +6 -0
index 3fe38f7315ed..bdfc0c17963f 100644
@@ -145,4 +145,10 @@ config PCI_MVEBU
 	  Say Y here if you want to enable PCIe controller support on
 	  Armada XP/38x SoCs.
 
+config PCI_KEYSTONE
+	bool "TI Keystone PCIe controller"
+	depends on DM_PCI
+	help
+	  Say Y here if you want to enable PCI controller support on AM654 SoC.
+
 endif
drivers/pci/Makefile +1 -0
index b5ebd50c855f..e54a98b8c9c5 100644
@@ -38,3 +38,4 @@ obj-$(CONFIG_PCIE_LAYERSCAPE_GEN4) += pcie_layerscape_gen4.o \
 				pcie_layerscape_gen4_fixup.o
 obj-$(CONFIG_PCI_XILINX) += pcie_xilinx.o
 obj-$(CONFIG_PCIE_INTEL_FPGA) += pcie_intel_fpga.o
+obj-$(CONFIG_PCI_KEYSTONE) += pcie_dw_ti.o
drivers/pci/pcie_dw_ti.c new file +725 -0

725 lines not on this page · raw message

Signature · 1 line
2.16.2
#12 lore raw

PATCH v3 3/7 pcie: ti: add driver for AM65x PCIe RC

Quoted 4 lines · Sekhar Nori

Add driver supporting PCIe root-complex available
on TI's AM65x SoC.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/3114939a/attachment.sig>
#13 lore raw

PATCH v3 4/7 phy: add support for AM654x SERDES

In reply to Sekhar Nori

Add a new SERDES driver for TI's AM654x SoC which configures the SERDES only for PCIe. Support fo USB3 can be added later.

SERDES in am654x has three input clocks (left input, external reference clock and right input) and two output clocks (left output and right output) in addition to a PLL mux clock which the SERDES uses for Clock Multiplier Unit (CMU refclock).

The PLL mux clock can select from one of the three input clocks. The right output can select between left input and external reference clock while the left output can select between the right input and external reference clock.

The driver has support to select PLL mux and left/right output mux as specified in device tree.

---
 drivers/phy/Kconfig        |   9 +
 drivers/phy/Makefile       |   1 +
 drivers/phy/phy-ti-am654.c | 411 +++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 421 insertions(+)
 create mode 100644 drivers/phy/phy-ti-am654.c
drivers/phy/Kconfig +9 -0
index 957efb3984bf..cce98448a4b5 100644
@@ -102,6 +102,15 @@ config SPL_PIPE3_PHY
 	  This PHY is found on omap devices supporting SATA such as dra7, am57x
 	  and omap5
 
+config AM654_PHY
+	tristate "TI AM654 SERDES support"
+	depends on PHY && ARCH_K3
+	select REGMAP
+	select SYSCON
+	help
+	  This option enables support for TI AM654 SerDes PHY used for
+	  PCIe.
+
 config STI_USB_PHY
 	bool "STMicroelectronics USB2 picoPHY driver for STiH407 family"
 	depends on PHY && ARCH_STI
drivers/phy/Makefile +1 -0
index 90646ca55b9d..39da13c4c941 100644
@@ -11,6 +11,7 @@ obj-$(CONFIG_BCM6358_USBH_PHY) += bcm6358-usbh-phy.o
 obj-$(CONFIG_BCM6368_USBH_PHY) += bcm6368-usbh-phy.o
 obj-$(CONFIG_PHY_SANDBOX) += sandbox-phy.o
 obj-$(CONFIG_$(SPL_)PIPE3_PHY) += ti-pipe3-phy.o
+obj-$(CONFIG_AM654_PHY) += phy-ti-am654.o
 obj-$(CONFIG_STI_USB_PHY) += sti_usb_phy.o
 obj-$(CONFIG_PHY_RCAR_GEN2) += phy-rcar-gen2.o
 obj-$(CONFIG_PHY_RCAR_GEN3) += phy-rcar-gen3.o
drivers/phy/phy-ti-am654.c new file +411 -0

411 lines not on this page · raw message

Signature · 1 line
2.16.2
#14 lore raw

PATCH v3 4/7 phy: add support for AM654x SERDES

Quoted 13 lines · Sekhar Nori

Add a new SERDES driver for TI's AM654x SoC which configures the SERDES only for PCIe. Support fo USB3 can be added later.

SERDES in am654x has three input clocks (left input, external reference clock and right input) and two output clocks (left output and right output) in addition to a PLL mux clock which the SERDES uses for Clock Multiplier Unit (CMU refclock).

The PLL mux clock can select from one of the three input clocks. The right output can select between left input and external reference clock while the left output can select between the right input and external reference clock.

Quoted 4 lines

The driver has support to select PLL mux and left/right output mux as specified in device tree.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/7f842ab5/attachment.sig>
#15 lore raw

PATCH v3 5/7 configs: am65x_evm_a53: enable PCIe support

In reply to Sekhar Nori

Enable support for PCIe and related configurations on AM654 EVM platform.

---
 configs/am65x_evm_a53_defconfig | 5 +++++
 1 file changed, 5 insertions(+)
configs/am65x_evm_a53_defconfig +5 -0
index 2cf3a693fc70..efb030549218 100644
@@ -35,6 +35,7 @@ CONFIG_CMD_ASKENV=y
 CONFIG_CMD_GPT=y
 CONFIG_CMD_I2C=y
 CONFIG_CMD_MMC=y
+CONFIG_CMD_PCI=y
 CONFIG_CMD_REMOTEPROC=y
 # CONFIG_CMD_SETEXPR is not set
 CONFIG_CMD_TIME=y
@@ -72,6 +73,10 @@ CONFIG_PHY_FIXED=y
 CONFIG_DM_ETH=y
 CONFIG_TI_AM65_CPSW_NUSS=y
 CONFIG_PHY=y
+CONFIG_PCI=y
+CONFIG_DM_PCI=y
+CONFIG_PCI_KEYSTONE=y
+CONFIG_AM654_PHY=y
 CONFIG_PINCTRL=y
 # CONFIG_PINCTRL_GENERIC is not set
 CONFIG_SPL_PINCTRL=y
Signature · 1 line
2.16.2
#16 lore raw

PATCH v3 5/7 configs: am65x_evm_a53: enable PCIe support

Quoted 4 lines · Sekhar Nori

Enable support for PCIe and related configurations on AM654 EVM platform.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/0ffed016/attachment.sig>
#17 lore raw

PATCH v3 6/7 arm: dts: k3-am65: add support for PCIe and SERDES

In reply to Sekhar Nori

Add needed device-tree nodes to support PCIe 0
and SERDES on AM65x SoC. The nodes are kept
disabled by default.

---
 arch/arm/dts/k3-am65-main.dtsi             | 108 +++++++++++++++++++++++++++++
 arch/arm/dts/k3-am65.dtsi                  |   1 +
 include/dt-bindings/phy/phy-am654-serdes.h |  13 ++++
 3 files changed, 122 insertions(+)
 create mode 100644 include/dt-bindings/phy/phy-am654-serdes.h
arch/arm/dts/k3-am65-main.dtsi +108 -0
index 7d0370605769..0f5da9a563d3 100644
@@ -5,6 +5,9 @@
  * Copyright (C) 2016-2018 Texas Instruments Incorporated - http://www.ti.com/
  */
 
+#include <dt-bindings/phy/phy-am654-serdes.h>
+#include <dt-bindings/phy/phy.h>
+
 &cbass_main {
 	gic500: interrupt-controller at 1800000 {
 		compatible = "arm,gic-v3";
@@ -143,4 +146,109 @@
 		clocks = <&k3_clks 113 1>;
 		power-domains = <&k3_pds 113 TI_SCI_PD_EXCLUSIVE>;
 	};
+
+	scm_conf: scm_conf at 100000 {
+		compatible = "syscon", "simple-mfd";
+		reg = <0 0x00100000 0 0x1c000>;
+		#address-cells = <1>;
+		#size-cells = <1>;
+		ranges = <0x0 0x0 0x00100000 0x1c000>;
+
+		serdes_mux: mux-controller {
+			compatible = "mmio-mux";
+			#mux-control-cells = <1>;
+			mux-reg-masks = <0x4080 0x3>, /* SERDES0 lane select */
+					<0x4090 0x3>; /* SERDES1 lane select */
+		};
+
+		pcie0_mode: pcie-mode at 4060 {
+			compatible = "syscon";
+			reg = <0x00004060 0x4>;
+		};
+
+		pcie1_mode: pcie-mode at 4070 {
+			compatible = "syscon";
+			reg = <0x00004070 0x4>;
+		};
+
+		serdes0_clk: serdes_clk at 4080 {
+			compatible = "syscon";
+			reg = <0x00004080 0x4>;
+		};
+
+		serdes1_clk: serdes_clk at 4090 {
+			compatible = "syscon";
+			reg = <0x00004090 0x4>;
+		};
+
+		pcie_devid: pcie-devid at 210 {
+			compatible = "syscon";
+			reg = <0x00000210 0x4>;
+		};
+	};
+
+	serdes0: serdes at 900000 {
+		compatible = "ti,phy-am654-serdes";
+		reg = <0x0 0x900000 0x0 0x2000>;
+		reg-names = "serdes";
+		#phy-cells = <2>;
+		power-domains = <&k3_pds 153 TI_SCI_PD_EXCLUSIVE>;
+		clocks = <&k3_clks 153 4>, <&k3_clks 153 1>, <&serdes1 AM654_SERDES_LO_REFCLK>;
+		clock-output-names = "serdes0_cmu_refclk", "serdes0_lo_refclk", "serdes0_ro_refclk";
+		assigned-clocks = <&k3_clks 153 4>, <&serdes0 AM654_SERDES_CMU_REFCLK>;
+		assigned-clock-parents = <&k3_clks 153 8>, <&k3_clks 153 4>;
+		ti,serdes-clk = <&serdes0_clk>;
+		mux-controls = <&serdes_mux 0>;
+		#clock-cells = <1>;
+	};
+
+	serdes1: serdes at 910000 {
+		compatible = "ti,phy-am654-serdes";
+		reg = <0x0 0x910000 0x0 0x2000>;
+		reg-names = "serdes";
+		#phy-cells = <2>;
+		power-domains = <&k3_pds 154 TI_SCI_PD_EXCLUSIVE>;
+		clocks = <&serdes0 AM654_SERDES_RO_REFCLK>, <&k3_clks 154 1>, <&k3_clks 154 5>;
+		clock-output-names = "serdes1_cmu_refclk", "serdes1_lo_refclk", "serdes1_ro_refclk";
+		assigned-clocks = <&k3_clks 154 5>, <&serdes1 AM654_SERDES_CMU_REFCLK>;
+		assigned-clock-parents = <&k3_clks 154 9>, <&k3_clks 154 5>;
+		ti,serdes-clk = <&serdes1_clk>;
+		mux-controls = <&serdes_mux 1>;
+		#clock-cells = <1>;
+	};
+
+	pcie0_rc: pcie at 5500000 {
+		compatible = "ti,am654-pcie-rc";
+		reg =  <0x0 0x5500000 0x0 0x1000>, <0x0 0x5501000 0x0 0x1000>, <0x0 0x10000000 0x0 0x2000>, <0x0 0x5506000 0x0 0x1000>;
+		reg-names = "app", "dbics", "config", "atu";
+		power-domains = <&k3_pds 120 TI_SCI_PD_EXCLUSIVE>;
+		#address-cells = <3>;
+		#size-cells = <2>;
+		ranges = <0x81000000 0 0          0x0   0x10020000 0 0x00010000
+			  0x82000000 0 0x10030000 0x0   0x10030000 0 0x07FD0000>;
+		ti,syscon-pcie-id = <&pcie_devid>;
+		ti,syscon-pcie-mode = <&pcie0_mode>;
+		bus-range = <0x0 0xff>;
+		status = "disabled";
+		device_type = "pci";
+		num-lanes = <1>;
+		num-ob-windows = <16>;
+		num-viewport = <16>;
+		max-link-speed = <3>;
+		interrupts = <GIC_SPI 340 IRQ_TYPE_EDGE_RISING>;
+		#interrupt-cells = <1>;
+		interrupt-map-mask = <0 0 0 7>;
+		interrupt-map = <0 0 0 1 &pcie0_intc 0>, /* INT A */
+				<0 0 0 2 &pcie0_intc 0>, /* INT B */
+				<0 0 0 3 &pcie0_intc 0>, /* INT C */
+				<0 0 0 4 &pcie0_intc 0>; /* INT D */
+		msi-map = <0x0 &gic_its 0x0 0x10000>;
+
+		pcie0_intc: legacy-interrupt-controller at 1 {
+			interrupt-controller;
+			#interrupt-cells = <1>;
+			interrupt-parent = <&gic500>;
+			interrupts = <GIC_SPI 328 IRQ_TYPE_EDGE_RISING>;
+		};
+	};
 };
arch/arm/dts/k3-am65.dtsi +1 -0
index a3abd146d104..a1467a4dd4d4 100644
@@ -69,6 +69,7 @@
 			 <0x00 0x00900000 0x00 0x00900000 0x00 0x00012000>, /* serdes */
 			 <0x00 0x01000000 0x00 0x01000000 0x00 0x0af02400>, /* Most peripherals */
 			 <0x00 0x30800000 0x00 0x30800000 0x00 0x0bc00000>, /* MAIN NAVSS */
+			 <0x00 0x10000000 0x00 0x10000000 0x00 0x10000000>, /* PCIe DAT */
 			 /* MCUSS Range */
 			 <0x00 0x28380000 0x00 0x28380000 0x00 0x03880000>,
 			 <0x00 0x40200000 0x00 0x40200000 0x00 0x00900100>,
include/dt-bindings/phy/phy-am654-serdes.h new file +13 -0
new file mode 100644
index 000000000000..e8d901729ed9
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * This header provides constants for AM654 SERDES.
+ */
+
+#ifndef _DT_BINDINGS_AM654_SERDES
+#define _DT_BINDINGS_AM654_SERDES
+
+#define AM654_SERDES_CMU_REFCLK	0
+#define AM654_SERDES_LO_REFCLK	1
+#define AM654_SERDES_RO_REFCLK	2
+
+#endif /* _DT_BINDINGS_AM654_SERDES */
Signature · 1 line
2.16.2
#18 lore raw

PATCH v3 6/7 arm: dts: k3-am65: add support for PCIe and SERDES

Quoted 5 lines · Sekhar Nori

Add needed device-tree nodes to support PCIe 0
and SERDES on AM65x SoC. The nodes are kept
disabled by default.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/d58eefcd/attachment.sig>
#19 lore raw

PATCH v3 7/7 configs: am65x_evm_a53: enable support for PCIe ethernet cards

In reply to Sekhar Nori

Enable support for Intel E1000 based PCIe ethernet cards that can be used to test PCIe RC functionality on AM65x EVM.

---
 configs/am65x_evm_a53_defconfig | 3 +++
 1 file changed, 3 insertions(+)
configs/am65x_evm_a53_defconfig +3 -0
index efb030549218..edc83675d8ae 100644
@@ -73,6 +73,9 @@ CONFIG_PHY_FIXED=y
 CONFIG_DM_ETH=y
 CONFIG_TI_AM65_CPSW_NUSS=y
 CONFIG_PHY=y
+CONFIG_DM_ETH=y
+CONFIG_E1000=y
+CONFIG_CMD_E1000=y
 CONFIG_PCI=y
 CONFIG_DM_PCI=y
 CONFIG_PCI_KEYSTONE=y
Signature · 1 line
2.16.2
#20 lore raw

PATCH v3 7/7 configs: am65x_evm_a53: enable support for PCIe ethernet cards

Quoted 4 lines · Sekhar Nori

Enable support for Intel E1000 based PCIe ethernet cards that can be used to test PCIe RC functionality on AM65x EVM.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190813/167f03ed/attachment.sig>