[U-Boot] [PATCH v2 00/21] dm: pci: Various fixes and test cases update

[U-Boot] [PATCH v2 00/21] dm: pci: Various fixes and test cases update

[Bin Meng]

This fixes several bugs in dm pci driver, as well as updating test
cases running on Sandbox to test dm pci.

It turns out adding test cases for v1 (3 commits) reveals more design
limitations/bugs of sandbox pci driver. Now the v2 series contains 20+
patches!

This series is available for testing at u-boot-x86/pci branch.

Changes in v2:
- Mask header_type with 0x7f

Bin Meng (21):
  pci: Remove 440ep specific macros
  dm: Correct typos in uclass_first/next_device_check()
  dm: core: Add ofnode function to read PCI vendor and device id
  dm: pci: Extract vendor/device id in child_post_bind()
  dm: pci: Fix scanning multi-function device
  test: dm: pci: Remove unnecessary steps in dm_test_pci_swapcase()
  test: dm: pci: Test more than one device on the same bus
  pci: sandbox: swap_case: Preserve space indicator bit in BAR registers
  test: dm: pci: Test more than one PCI host controller
  test: dm: pci: Add tests for configuration space access
  pci: sandbox: emul: Fix the call to pci_bus_find_devfn()
  dm: pci: Assign correct driver data when binding a driver
  pci: sandbox: Support dynamically binding device driver
  pci: sandbox: swap_case: Declare dynamic driver matching
  sandbox: Update test.dts for dynamic PCI device driver matching
  test: dm: pci: Test driver binding with driver data provided
  pci: sandbox: emul: Rename priv structure
  test: dm: pci: Add tests for mixed static and dynamic devices on the
    same bus
  pci: Add all known capability and extended capability ids
  dm: pci: Add APIs to find capability and extended capability
  test: dm: pci: Add cases for finding PCI capability APIs

 arch/sandbox/dts/test.dts       |  41 +++++++++-
 arch/sandbox/include/asm/test.h |  12 +++
 doc/driver-model/pci-info.txt   |  23 ++++++
 drivers/core/ofnode.c           |  36 +++++++++
 drivers/misc/swap_case.c        |  36 ++++++++-
 drivers/pci/pci-emul-uclass.c   |  24 +++---
 drivers/pci/pci-uclass.c        |  83 ++++++++++++++++++--
 drivers/pci/pci_sandbox.c       |  78 ++++++++++++++++--
 include/dm/ofnode.h             |  13 +++
 include/dm/uclass.h             |   4 +-
 include/pci.h                   |  84 ++++++++++++++++----
 test/dm/pci.c                   | 170 ++++++++++++++++++++++++++++++++++++++--
 12 files changed, 551 insertions(+), 53 deletions(-)

-- 
2.7.4

[U-Boot] [PATCH v2 01/21] pci: Remove 440ep specific macros

[Bin Meng]

These macros should not be put in the generic pci.h header file.
Since they are not referenced anywhere, remove them completely.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 include/pci.h | 15 ---------------
 1 file changed, 15 deletions(-)

diff --git a/include/pci.h b/include/pci.h
index 8e27cbf..427094c 100644
--- a/include/pci.h
+++ b/include/pci.h
@@ -271,21 +271,6 @@
 #define  PCI_BRIDGE_CTL_BUS_RESET 0x40	/* Secondary bus reset */
 #define  PCI_BRIDGE_CTL_FAST_BACK 0x80	/* Fast Back2Back enabled on secondary interface */
 
-/* From 440ep */
-#define PCI_ERREN       0x48     /* Error Enable */
-#define PCI_ERRSTS      0x49     /* Error Status */
-#define PCI_BRDGOPT1    0x4A     /* PCI Bridge Options 1 */
-#define PCI_PLBSESR0    0x4C     /* PCI PLB Slave Error Syndrome 0 */
-#define PCI_PLBSESR1    0x50     /* PCI PLB Slave Error Syndrome 1 */
-#define PCI_PLBSEAR     0x54     /* PCI PLB Slave Error Address */
-#define PCI_CAPID       0x58     /* Capability Identifier */
-#define PCI_NEXTITEMPTR 0x59     /* Next Item Pointer */
-#define PCI_PMC         0x5A     /* Power Management Capabilities */
-#define PCI_PMCSR       0x5C     /* Power Management Control Status */
-#define PCI_PMCSRBSE    0x5E     /* PMCSR PCI to PCI Bridge Support Extensions */
-#define PCI_BRDGOPT2    0x60     /* PCI Bridge Options 2 */
-#define PCI_PMSCRR      0x64     /* Power Management State Change Request Re. */
-
 /* Header type 2 (CardBus bridges) */
 #define PCI_CB_CAPABILITY_LIST	0x14
 /* 0x15 reserved */
-- 
2.7.4

[U-Boot] [PATCH v2 02/21] dm: Correct typos in uclass_first/next_device_check()

[Bin Meng]

Correct typos in the comment block of uclass_first/next_device_check().

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 include/dm/uclass.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/dm/uclass.h b/include/dm/uclass.h
index 9fbaa7d..0e882ce 100644
--- a/include/dm/uclass.h
+++ b/include/dm/uclass.h
@@ -302,7 +302,7 @@ int uclass_first_device_err(enum uclass_id id, struct udevice **devp);
 int uclass_next_device(struct udevice **devp);
 
 /**
- * uclass_first_device() - Get the first device in a uclass
+ * uclass_first_device_check() - Get the first device in a uclass
  *
  * The device returned is probed if necessary, and ready for use
  *
@@ -318,7 +318,7 @@ int uclass_next_device(struct udevice **devp);
 int uclass_first_device_check(enum uclass_id id, struct udevice **devp);
 
 /**
- * uclass_next_device() - Get the next device in a uclass
+ * uclass_next_device_check() - Get the next device in a uclass
  *
  * The device returned is probed if necessary, and ready for use
  *
-- 
2.7.4

[U-Boot] [PATCH v2 03/21] dm: core: Add ofnode function to read PCI vendor and device id

[Bin Meng]

We don't have the live-tree version of fdtdec_get_pci_vendev().
This adds the API.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 drivers/core/ofnode.c | 36 ++++++++++++++++++++++++++++++++++++
 include/dm/ofnode.h   | 13 +++++++++++++
 2 files changed, 49 insertions(+)

diff --git a/drivers/core/ofnode.c b/drivers/core/ofnode.c
index 2937539..0cfb0fb 100644
--- a/drivers/core/ofnode.c
+++ b/drivers/core/ofnode.c
@@ -623,6 +623,42 @@ fail:
 	return ret;
 }
 
+int ofnode_read_pci_vendev(ofnode node, u16 *vendor, u16 *device)
+{
+	const char *list, *end;
+	int len;
+
+	list = ofnode_get_property(node, "compatible", &len);
+	if (!list)
+		return -ENOENT;
+
+	end = list + len;
+	while (list < end) {
+		len = strlen(list);
+		if (len >= strlen("pciVVVV,DDDD")) {
+			char *s = strstr(list, "pci");
+
+			/*
+			 * check if the string is something like pciVVVV,DDDD.RR
+			 * or just pciVVVV,DDDD
+			 */
+			if (s && s[7] == ',' &&
+			    (s[12] == '.' || s[12] == 0)) {
+				s += 3;
+				*vendor = simple_strtol(s, NULL, 16);
+
+				s += 5;
+				*device = simple_strtol(s, NULL, 16);
+
+				return 0;
+			}
+		}
+		list += (len + 1);
+	}
+
+	return -ENOENT;
+}
+
 int ofnode_read_addr_cells(ofnode node)
 {
 	if (ofnode_is_np(node))
diff --git a/include/dm/ofnode.h b/include/dm/ofnode.h
index cd08a7e..ab36b74 100644
--- a/include/dm/ofnode.h
+++ b/include/dm/ofnode.h
@@ -586,6 +586,19 @@ int ofnode_read_pci_addr(ofnode node, enum fdt_pci_space type,
 			 const char *propname, struct fdt_pci_addr *addr);
 
 /**
+ * ofnode_read_pci_vendev() - look up PCI vendor and device id
+ *
+ * Look at the compatible property of a device node that represents a PCI
+ * device and extract pci vendor id and device id from it.
+ *
+ * @param node		node to examine
+ * @param vendor	vendor id of the pci device
+ * @param device	device id of the pci device
+ * @return 0 if ok, negative on error
+ */
+int ofnode_read_pci_vendev(ofnode node, u16 *vendor, u16 *device);
+
+/**
  * ofnode_read_addr_cells() - Get the number of address cells for a node
  *
  * This walks back up the tree to find the closest #address-cells property
-- 
2.7.4

[U-Boot] [PATCH v2 04/21] dm: pci: Extract vendor/device id in child_post_bind()

[Bin Meng]

Currently only devfn is extracted in child_post_bind(). Now that
we have the live-tree version API to look up PCI vendor and device
id from the compatible string, let's extract and save them too.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 drivers/pci/pci-uclass.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c
index 46e9c71..b2909a3 100644
--- a/drivers/pci/pci-uclass.c
+++ b/drivers/pci/pci-uclass.c
@@ -987,19 +987,18 @@ static int pci_uclass_child_post_bind(struct udevice *dev)
 	if (!dev_of_valid(dev))
 		return 0;
 
-	/*
-	 * We could read vendor, device, class if available. But for now we
-	 * just check the address.
-	 */
 	pplat = dev_get_parent_platdata(dev);
+
+	/* Extract vendor id and device id if available */
+	ofnode_read_pci_vendev(dev_ofnode(dev), &pplat->vendor, &pplat->device);
+
+	/* Extract the devfn from fdt_pci_addr */
 	ret = ofnode_read_pci_addr(dev_ofnode(dev), FDT_PCI_SPACE_CONFIG, "reg",
 				   &addr);
-
 	if (ret) {
 		if (ret != -ENOENT)
 			return -EINVAL;
 	} else {
-		/* extract the devfn from fdt_pci_addr */
 		pplat->devfn = addr.phys_hi & 0xff00;
 	}
 
-- 
2.7.4

[U-Boot] [PATCH v2 05/21] dm: pci: Fix scanning multi-function device

[Bin Meng]

The flag to control whether to scan multi-function device during
enumeration should be cleared at the beginning of each iteration
if the device's function number equals to zero.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 drivers/pci/pci-uclass.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c
index b2909a3..28ea8c5 100644
--- a/drivers/pci/pci-uclass.c
+++ b/drivers/pci/pci-uclass.c
@@ -745,6 +745,8 @@ int pci_bind_bus_devices(struct udevice *bus)
 		struct udevice *dev;
 		ulong class;
 
+		if (!PCI_FUNC(bdf))
+			found_multi = false;
 		if (PCI_FUNC(bdf) && !found_multi)
 			continue;
 		/* Check only the first access, we don't expect problems */
-- 
2.7.4

[U-Boot] [PATCH v2 06/21] test: dm: pci: Remove unnecessary steps in dm_test_pci_swapcase()

[Bin Meng]

The check on uclass_get_device() and device_active() is unnecessary
as the follow-up test operations will implicitly probe the driver.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 test/dm/pci.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/test/dm/pci.c b/test/dm/pci.c
index 47b5d22..be1208c 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -34,14 +34,12 @@ DM_TEST(dm_test_pci_busnum, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
 /* Test that we can use the swapcase device correctly */
 static int dm_test_pci_swapcase(struct unit_test_state *uts)
 {
-	struct udevice *emul, *swap;
+	struct udevice *swap;
 	ulong io_addr, mem_addr;
 	char *ptr;
 
 	/* Check that asking for the device automatically fires up PCI */
-	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 0, &emul));
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x1f, 0), &swap));
-	ut_assert(device_active(swap));
 
 	/* First test I/O */
 	io_addr = dm_pci_read_bar32(swap, 0);
-- 
2.7.4

[U-Boot] [PATCH v2 07/21] test: dm: pci: Test more than one device on the same bus

[Bin Meng]

It's quite common to have more than one device on the same PCI bus.
This updates the test case to test such scenario.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 arch/sandbox/dts/test.dts |  7 +++++++
 test/dm/pci.c             | 37 +++++++++++++++++++++++++++++++++----
 2 files changed, 40 insertions(+), 4 deletions(-)

diff --git a/arch/sandbox/dts/test.dts b/arch/sandbox/dts/test.dts
index 137679a..237266d 100644
--- a/arch/sandbox/dts/test.dts
+++ b/arch/sandbox/dts/test.dts
@@ -302,6 +302,13 @@
 		#size-cells = <2>;
 		ranges = <0x02000000 0 0x10000000 0x10000000 0 0x2000
 				0x01000000 0 0x20000000 0x20000000 0 0x2000>;
+		pci at 0,0 {
+			compatible = "pci-generic";
+			reg = <0x0000 0 0 0 0>;
+			emul at 0,0 {
+				compatible = "sandbox,swap-case";
+			};
+		};
 		pci at 1f,0 {
 			compatible = "pci-generic";
 			reg = <0xf800 0 0 0 0>;
diff --git a/test/dm/pci.c b/test/dm/pci.c
index be1208c..f2bd52a 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -20,16 +20,24 @@ static int dm_test_pci_base(struct unit_test_state *uts)
 }
 DM_TEST(dm_test_pci_base, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
 
-/* Test that sandbox PCI bus numbering works correctly */
-static int dm_test_pci_busnum(struct unit_test_state *uts)
+/* Test that sandbox PCI bus numbering and device works correctly */
+static int dm_test_pci_busdev(struct unit_test_state *uts)
 {
 	struct udevice *bus;
+	struct udevice *emul, *swap;
 
 	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 0, &bus));
 
+	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 0, &emul));
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x00, 0), &swap));
+	ut_assert(device_active(swap));
+	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 1, &emul));
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x1f, 0), &swap));
+	ut_assert(device_active(swap));
+
 	return 0;
 }
-DM_TEST(dm_test_pci_busnum, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
+DM_TEST(dm_test_pci_busdev, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
 
 /* Test that we can use the swapcase device correctly */
 static int dm_test_pci_swapcase(struct unit_test_state *uts)
@@ -38,7 +46,28 @@ static int dm_test_pci_swapcase(struct unit_test_state *uts)
 	ulong io_addr, mem_addr;
 	char *ptr;
 
-	/* Check that asking for the device automatically fires up PCI */
+	/* Check that asking for the device 0 automatically fires up PCI */
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x00, 0), &swap));
+
+	/* First test I/O */
+	io_addr = dm_pci_read_bar32(swap, 0);
+	outb(2, io_addr);
+	ut_asserteq(2, inb(io_addr));
+
+	/*
+	 * Now test memory mapping - note we must unmap and remap to cause
+	 * the swapcase emulation to see our data and response.
+	 */
+	mem_addr = dm_pci_read_bar32(swap, 1);
+	ptr = map_sysmem(mem_addr, 20);
+	strcpy(ptr, "This is a TesT");
+	unmap_sysmem(ptr);
+
+	ptr = map_sysmem(mem_addr, 20);
+	ut_asserteq_str("tHIS IS A tESt", ptr);
+	unmap_sysmem(ptr);
+
+	/* Check that asking for the device 1 automatically fires up PCI */
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x1f, 0), &swap));
 
 	/* First test I/O */
-- 
2.7.4

[U-Boot] [PATCH v2 08/21] pci: sandbox: swap_case: Preserve space indicator bit in BAR registers

[Bin Meng]

With the newly added testing of more than one device, we get:

  => ut dm pci_swapcase
  Test: dm_test_pci_swapcase: pci.c
  test/dm/pci.c:88, dm_test_pci_swapcase(): "tHIS IS A tESt" = ptr:
  Expected "tHIS IS A tESt", got "this is a test"
  Test: dm_test_pci_swapcase: pci.c (flat tree)
  test/dm/pci.c:88, dm_test_pci_swapcase(): "tHIS IS A tESt" = ptr:
  Expected "tHIS IS A tESt", got "this is a test"
  Failures: 2

The failure only happens on the 2nd swap_case device on the PCI bus.
The case passes on the 1st device.

It turns out the swap_case driver does not emulate bit#0 in BAR
registers as a read-only bit. This corrects the implementation.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 drivers/misc/swap_case.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/misc/swap_case.c b/drivers/misc/swap_case.c
index b777404..80ccb9f 100644
--- a/drivers/misc/swap_case.c
+++ b/drivers/misc/swap_case.c
@@ -142,6 +142,8 @@ static int sandbox_swap_case_write_config(struct udevice *emul, uint offset,
 
 		debug("w bar %d=%lx\n", barnum, value);
 		*bar = value;
+		/* space indicator (bit#0) is read-only */
+		*bar |= barinfo[barnum].type;
 		break;
 	}
 	}
@@ -157,11 +159,11 @@ static int sandbox_swap_case_find_bar(struct udevice *emul, unsigned int addr,
 
 	for (barnum = 0; barnum < ARRAY_SIZE(barinfo); barnum++) {
 		unsigned int size = barinfo[barnum].size;
+		u32 base = plat->bar[barnum] & ~PCI_BASE_ADDRESS_SPACE;
 
-		if (addr >= plat->bar[barnum] &&
-		    addr < plat->bar[barnum] + size) {
+		if (addr >= base && addr < base + size) {
 			*barnump = barnum;
-			*offsetp = addr - plat->bar[barnum];
+			*offsetp = addr - base;
 			return 0;
 		}
 	}
-- 
2.7.4

[U-Boot] [PATCH v2 09/21] test: dm: pci: Test more than one PCI host controller

[Bin Meng]

So far there is only one PCI host controller in the sandbox test
configuration. This is normally the case for x86, but it can be
common on other architectures like ARM/PPC to have more than one
PCI host controller in the system.

This updates the case to cover such scenario.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 arch/sandbox/dts/test.dts | 28 ++++++++++++++++++++++++++--
 test/dm/pci.c             | 11 +++++++++++
 2 files changed, 37 insertions(+), 2 deletions(-)

diff --git a/arch/sandbox/dts/test.dts b/arch/sandbox/dts/test.dts
index 237266d..0bce6d0 100644
--- a/arch/sandbox/dts/test.dts
+++ b/arch/sandbox/dts/test.dts
@@ -14,7 +14,8 @@
 		i2c0 = "/i2c at 0";
 		mmc0 = "/mmc0";
 		mmc1 = "/mmc1";
-		pci0 = &pci;
+		pci0 = &pci0;
+		pci1 = &pci1;
 		remoteproc1 = &rproc_1;
 		remoteproc2 = &rproc_2;
 		rtc0 = &rtc_0;
@@ -295,7 +296,7 @@
 		compatible = "sandbox,mmc";
 	};
 
-	pci: pci-controller {
+	pci0: pci-controller0 {
 		compatible = "sandbox,pci";
 		device_type = "pci";
 		#address-cells = <3>;
@@ -318,6 +319,29 @@
 		};
 	};
 
+	pci1: pci-controller1 {
+		compatible = "sandbox,pci";
+		device_type = "pci";
+		#address-cells = <3>;
+		#size-cells = <2>;
+		ranges = <0x02000000 0 0x30000000 0x30000000 0 0x2000
+				0x01000000 0 0x40000000 0x40000000 0 0x2000>;
+		pci at 8,0 {
+			compatible = "pci-generic";
+			reg = <0x4000 0 0 0 0>;
+			emul at 8,0 {
+				compatible = "sandbox,swap-case";
+			};
+		};
+		pci at c,0 {
+			compatible = "pci-generic";
+			reg = <0x6000 0 0 0 0>;
+			emul at c,0 {
+				compatible = "sandbox,swap-case";
+			};
+		};
+	};
+
 	probing {
 		compatible = "simple-bus";
 		test1 {
diff --git a/test/dm/pci.c b/test/dm/pci.c
index f2bd52a..727ec34 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -26,6 +26,7 @@ static int dm_test_pci_busdev(struct unit_test_state *uts)
 	struct udevice *bus;
 	struct udevice *emul, *swap;
 
+	/* Test bus#0 and its devices */
 	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 0, &bus));
 
 	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 0, &emul));
@@ -35,6 +36,16 @@ static int dm_test_pci_busdev(struct unit_test_state *uts)
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x1f, 0), &swap));
 	ut_assert(device_active(swap));
 
+	/* Test bus#1 and its devices */
+	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 1, &bus));
+
+	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 2, &emul));
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x08, 0), &swap));
+	ut_assert(device_active(swap));
+	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 3, &emul));
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x0c, 0), &swap));
+	ut_assert(device_active(swap));
+
 	return 0;
 }
 DM_TEST(dm_test_pci_busdev, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
-- 
2.7.4

[U-Boot] [PATCH v2 10/21] test: dm: pci: Add tests for configuration space access

[Bin Meng]

So far we missed the testing for PCI configuration space access.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 test/dm/pci.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/test/dm/pci.c b/test/dm/pci.c
index 727ec34..089b72e 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -6,6 +6,7 @@
 #include <common.h>
 #include <dm.h>
 #include <asm/io.h>
+#include <asm/test.h>
 #include <dm/test.h>
 #include <test/ut.h>
 
@@ -24,27 +25,32 @@ DM_TEST(dm_test_pci_base, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
 static int dm_test_pci_busdev(struct unit_test_state *uts)
 {
 	struct udevice *bus;
-	struct udevice *emul, *swap;
+	struct udevice *swap;
+	u16 vendor, device;
 
 	/* Test bus#0 and its devices */
 	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 0, &bus));
 
-	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 0, &emul));
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x00, 0), &swap));
-	ut_assert(device_active(swap));
-	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 1, &emul));
+	vendor = 0;
+	ut_assertok(dm_pci_read_config16(swap, PCI_VENDOR_ID, &vendor));
+	ut_asserteq(SANDBOX_PCI_VENDOR_ID, vendor);
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x1f, 0), &swap));
-	ut_assert(device_active(swap));
+	device = 0;
+	ut_assertok(dm_pci_read_config16(swap, PCI_DEVICE_ID, &device));
+	ut_asserteq(SANDBOX_PCI_DEVICE_ID, device);
 
 	/* Test bus#1 and its devices */
 	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 1, &bus));
 
-	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 2, &emul));
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x08, 0), &swap));
-	ut_assert(device_active(swap));
-	ut_assertok(uclass_get_device(UCLASS_PCI_EMUL, 3, &emul));
+	vendor = 0;
+	ut_assertok(dm_pci_read_config16(swap, PCI_VENDOR_ID, &vendor));
+	ut_asserteq(SANDBOX_PCI_VENDOR_ID, vendor);
 	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x0c, 0), &swap));
-	ut_assert(device_active(swap));
+	device = 0;
+	ut_assertok(dm_pci_read_config16(swap, PCI_DEVICE_ID, &device));
+	ut_asserteq(SANDBOX_PCI_DEVICE_ID, device);
 
 	return 0;
 }
-- 
2.7.4

[U-Boot] [PATCH v2 11/21] pci: sandbox: emul: Fix the call to pci_bus_find_devfn()

[Bin Meng]

With the newly added test cases for PCI configuration access, we get:

  => ut dm pci_busdev
  Test: dm_test_pci_busdev: pci.c
  test/dm/pci.c:49, dm_test_pci_busdev(): SANDBOX_PCI_VENDOR_ID == vendor:
  Expected 4660, got 65535
  Test: dm_test_pci_busdev: pci.c (flat tree)
  test/dm/pci.c:49, dm_test_pci_busdev(): SANDBOX_PCI_VENDOR_ID == vendor:
  Expected 4660, got 65535
  Failures: 2

The bug only shows up when bus number is not equal to zero. This is
caused by the plain find_devfn parameter is passed to function call
pci_bus_find_devfn(), inside which find_devfn is compared to devfn
in the device's pplat structure. However pplat->devfn does not carry
the bus number. Fix this by passing find_devfn with bus number masked.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 drivers/pci/pci-emul-uclass.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/pci-emul-uclass.c b/drivers/pci/pci-emul-uclass.c
index 79e2c14..8570a5d 100644
--- a/drivers/pci/pci-emul-uclass.c
+++ b/drivers/pci/pci-emul-uclass.c
@@ -21,7 +21,7 @@ int sandbox_pci_get_emul(struct udevice *bus, pci_dev_t find_devfn,
 	struct udevice *dev;
 	int ret;
 
-	ret = pci_bus_find_devfn(bus, find_devfn, &dev);
+	ret = pci_bus_find_devfn(bus, PCI_MASK_BUS(find_devfn), &dev);
 	if (ret) {
 		debug("%s: Could not find emulator for dev %x\n", __func__,
 		      find_devfn);
-- 
2.7.4

[U-Boot] [PATCH v2 12/21] dm: pci: Assign correct driver data when binding a driver

[Bin Meng]

The correct driver data comes from the matching 'id' instead of
'find_id' in pci_find_and_bind_driver().

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
---

Changes in v2: None

 drivers/pci/pci-uclass.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c
index 28ea8c5..5eb6841 100644
--- a/drivers/pci/pci-uclass.c
+++ b/drivers/pci/pci-uclass.c
@@ -690,7 +690,7 @@ static int pci_find_and_bind_driver(struct udevice *parent,
 			if (ret)
 				goto error;
 			debug("%s: Match found: %s\n", __func__, drv->name);
-			dev->driver_data = find_id->driver_data;
+			dev->driver_data = id->driver_data;
 			*devp = dev;
 			return 0;
 		}
-- 
2.7.4

[U-Boot] [PATCH v2 13/21] pci: sandbox: Support dynamically binding device driver

[Bin Meng]

At present all emulated sandbox pci devices must be present in the
device tree in order to be used. The real world pci uclass driver
supports pci device driver matching, and we should add such support
on sandbox too.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 doc/driver-model/pci-info.txt | 23 +++++++++++++
 drivers/pci/pci-emul-uclass.c | 14 +++++---
 drivers/pci/pci_sandbox.c     | 78 +++++++++++++++++++++++++++++++++++++++----
 include/pci.h                 |  3 +-
 4 files changed, 107 insertions(+), 11 deletions(-)

diff --git a/doc/driver-model/pci-info.txt b/doc/driver-model/pci-info.txt
index 52b4389..e67264b 100644
--- a/doc/driver-model/pci-info.txt
+++ b/doc/driver-model/pci-info.txt
@@ -133,3 +133,26 @@ When this bus is scanned we will end up with something like this:
 
 When accesses go to the pci at 1f,0 device they are forwarded to its child, the
 emulator.
+
+The sandbox PCI drivers also support dynamic driver binding, allowing device
+driver to declare the driver binding information via U_BOOT_PCI_DEVICE(),
+eliminating the need to provide any device tree node under the host controller
+node. It is required a "sandbox,dev-info" property must be provided in the
+host controller node for this functionality to work.
+
+	pci1: pci-controller1 {
+		compatible = "sandbox,pci";
+		...
+		sandbox,dev-info = <0x08 0x00 0x1234 0x5678
+				    0x0c 0x00 0x1234 0x5678>;
+	};
+
+When this bus is scanned we will end up with something like this:
+
+ pci        [ + ]   pci_sandbo  |-- pci-controller1
+ pci_emul   [   ]   sandbox_sw  |   |-- sandbox_swap_case_emul
+ pci_emul   [   ]   sandbox_sw  |   `-- sandbox_swap_case_emul
+
+Note the difference from the statically declared device nodes is that the
+device is directly attached to the host controller, instead of via a container
+device like pci at 1f,0.
diff --git a/drivers/pci/pci-emul-uclass.c b/drivers/pci/pci-emul-uclass.c
index 8570a5d..e9d2f49 100644
--- a/drivers/pci/pci-emul-uclass.c
+++ b/drivers/pci/pci-emul-uclass.c
@@ -16,21 +16,27 @@ struct sandbox_pci_priv {
 };
 
 int sandbox_pci_get_emul(struct udevice *bus, pci_dev_t find_devfn,
-			 struct udevice **emulp)
+			 struct udevice **containerp, struct udevice **emulp)
 {
 	struct udevice *dev;
 	int ret;
 
+	*containerp = NULL;
 	ret = pci_bus_find_devfn(bus, PCI_MASK_BUS(find_devfn), &dev);
 	if (ret) {
 		debug("%s: Could not find emulator for dev %x\n", __func__,
 		      find_devfn);
 		return ret;
 	}
+	*containerp = dev;
 
-	ret = device_find_first_child(dev, emulp);
-	if (ret)
-		return ret;
+	if (device_get_uclass_id(dev) == UCLASS_PCI_GENERIC) {
+		ret = device_find_first_child(dev, emulp);
+		if (ret)
+			return ret;
+	} else {
+		*emulp = dev;
+	}
 
 	return *emulp ? 0 : -ENODEV;
 }
diff --git a/drivers/pci/pci_sandbox.c b/drivers/pci/pci_sandbox.c
index 67cd733..119a98d 100644
--- a/drivers/pci/pci_sandbox.c
+++ b/drivers/pci/pci_sandbox.c
@@ -10,15 +10,27 @@
 #include <inttypes.h>
 #include <pci.h>
 
+#define FDT_DEV_INFO_CELLS	4
+#define FDT_DEV_INFO_SIZE	(FDT_DEV_INFO_CELLS * sizeof(u32))
+
+#define SANDBOX_PCI_DEVFN(d, f)	((d << 3) | f)
+
+struct sandbox_pci_priv {
+	struct {
+		u16 vendor;
+		u16 device;
+	} vendev[256];
+};
+
 static int sandbox_pci_write_config(struct udevice *bus, pci_dev_t devfn,
 				    uint offset, ulong value,
 				    enum pci_size_t size)
 {
 	struct dm_pci_emul_ops *ops;
-	struct udevice *emul;
+	struct udevice *container, *emul;
 	int ret;
 
-	ret = sandbox_pci_get_emul(bus, devfn, &emul);
+	ret = sandbox_pci_get_emul(bus, devfn, &container, &emul);
 	if (ret)
 		return ret == -ENODEV ? 0 : ret;
 	ops = pci_get_emul_ops(emul);
@@ -33,14 +45,31 @@ static int sandbox_pci_read_config(struct udevice *bus, pci_dev_t devfn,
 				   enum pci_size_t size)
 {
 	struct dm_pci_emul_ops *ops;
-	struct udevice *emul;
+	struct udevice *container, *emul;
+	struct sandbox_pci_priv *priv = dev_get_priv(bus);
 	int ret;
 
 	/* Prepare the default response */
 	*valuep = pci_get_ff(size);
-	ret = sandbox_pci_get_emul(bus, devfn, &emul);
-	if (ret)
-		return ret == -ENODEV ? 0 : ret;
+	ret = sandbox_pci_get_emul(bus, devfn, &container, &emul);
+	if (ret) {
+		if (!container) {
+			u16 vendor, device;
+
+			devfn = SANDBOX_PCI_DEVFN(PCI_DEV(devfn),
+						  PCI_FUNC(devfn));
+			vendor = priv->vendev[devfn].vendor;
+			device = priv->vendev[devfn].device;
+			if (offset == PCI_VENDOR_ID && vendor)
+				*valuep = vendor;
+			else if (offset == PCI_DEVICE_ID && device)
+				*valuep = device;
+
+			return 0;
+		} else {
+			return ret == -ENODEV ? 0 : ret;
+		}
+	}
 	ops = pci_get_emul_ops(emul);
 	if (!ops || !ops->read_config)
 		return -ENOSYS;
@@ -48,6 +77,41 @@ static int sandbox_pci_read_config(struct udevice *bus, pci_dev_t devfn,
 	return ops->read_config(emul, offset, valuep, size);
 }
 
+static int sandbox_pci_probe(struct udevice *dev)
+{
+	struct sandbox_pci_priv *priv = dev_get_priv(dev);
+	const fdt32_t *cell;
+	u8 pdev, pfn, devfn;
+	int len;
+
+	cell = ofnode_get_property(dev_ofnode(dev), "sandbox,dev-info", &len);
+	if (!cell)
+		return 0;
+
+	if ((len % FDT_DEV_INFO_SIZE) == 0) {
+		int num = len / FDT_DEV_INFO_SIZE;
+		int i;
+
+		for (i = 0; i < num; i++) {
+			debug("dev info #%d: %02x %02x %04x %04x\n", i,
+			      fdt32_to_cpu(cell[0]), fdt32_to_cpu(cell[1]),
+			      fdt32_to_cpu(cell[2]), fdt32_to_cpu(cell[3]));
+
+			pdev = fdt32_to_cpu(cell[0]);
+			pfn = fdt32_to_cpu(cell[1]);
+			if (pdev > 31 || pfn > 7)
+				continue;
+			devfn = SANDBOX_PCI_DEVFN(pdev, pfn);
+			priv->vendev[devfn].vendor = fdt32_to_cpu(cell[2]);
+			priv->vendev[devfn].device = fdt32_to_cpu(cell[3]);
+
+			cell += FDT_DEV_INFO_CELLS;
+		}
+	}
+
+	return 0;
+}
+
 static const struct dm_pci_ops sandbox_pci_ops = {
 	.read_config = sandbox_pci_read_config,
 	.write_config = sandbox_pci_write_config,
@@ -63,6 +127,8 @@ U_BOOT_DRIVER(pci_sandbox) = {
 	.id	= UCLASS_PCI,
 	.of_match = sandbox_pci_ids,
 	.ops	= &sandbox_pci_ops,
+	.probe	= sandbox_pci_probe,
+	.priv_auto_alloc_size = sizeof(struct sandbox_pci_priv),
 
 	/* Attach an emulator if we can */
 	.child_post_bind = dm_scan_fdt_dev,
diff --git a/include/pci.h b/include/pci.h
index 427094c..c0ae5d1 100644
--- a/include/pci.h
+++ b/include/pci.h
@@ -1441,11 +1441,12 @@ struct dm_pci_emul_ops {
  *
  * @bus:	PCI bus to search
  * @find_devfn:	PCI device and function address (PCI_DEVFN())
+ * @containerp:	Returns container device if found
  * @emulp:	Returns emulated device if found
  * @return 0 if found, -ENODEV if not found
  */
 int sandbox_pci_get_emul(struct udevice *bus, pci_dev_t find_devfn,
-			 struct udevice **emulp);
+			 struct udevice **containerp, struct udevice **emulp);
 
 #endif /* CONFIG_DM_PCI */
 
-- 
2.7.4

[U-Boot] [PATCH v2 14/21] pci: sandbox: swap_case: Declare dynamic driver matching

[Bin Meng]

This adds a U_BOOT_PCI_DEVICE() declaration to the swap_case driver.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 arch/sandbox/include/asm/test.h | 4 ++++
 drivers/misc/swap_case.c        | 7 +++++++
 2 files changed, 11 insertions(+)

diff --git a/arch/sandbox/include/asm/test.h b/arch/sandbox/include/asm/test.h
index 08863bf..57aeca8 100644
--- a/arch/sandbox/include/asm/test.h
+++ b/arch/sandbox/include/asm/test.h
@@ -16,6 +16,10 @@
 #define SANDBOX_PCI_CLASS_CODE		PCI_CLASS_CODE_COMM
 #define SANDBOX_PCI_CLASS_SUB_CODE	PCI_CLASS_SUB_CODE_COMM_SERIAL
 
+/* Useful for PCI_VDEVICE() macro */
+#define PCI_VENDOR_ID_SANDBOX		SANDBOX_PCI_VENDOR_ID
+#define SWAP_CASE_DRV_DATA		0x55aa
+
 #define SANDBOX_CLK_RATE		32768
 
 /* System controller driver data */
diff --git a/drivers/misc/swap_case.c b/drivers/misc/swap_case.c
index 80ccb9f..790bb0c 100644
--- a/drivers/misc/swap_case.c
+++ b/drivers/misc/swap_case.c
@@ -285,3 +285,10 @@ U_BOOT_DRIVER(sandbox_swap_case_emul) = {
 	.priv_auto_alloc_size = sizeof(struct swap_case_priv),
 	.platdata_auto_alloc_size = sizeof(struct swap_case_platdata),
 };
+
+static struct pci_device_id sandbox_swap_case_supported[] = {
+	{ PCI_VDEVICE(SANDBOX, SANDBOX_PCI_DEVICE_ID), SWAP_CASE_DRV_DATA },
+	{},
+};
+
+U_BOOT_PCI_DEVICE(sandbox_swap_case_emul, sandbox_swap_case_supported);
-- 
2.7.4

[U-Boot] [PATCH v2 15/21] sandbox: Update test.dts for dynamic PCI device driver matching

[Bin Meng]

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 arch/sandbox/dts/test.dts | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/arch/sandbox/dts/test.dts b/arch/sandbox/dts/test.dts
index 0bce6d0..44d24f9 100644
--- a/arch/sandbox/dts/test.dts
+++ b/arch/sandbox/dts/test.dts
@@ -326,20 +326,8 @@
 		#size-cells = <2>;
 		ranges = <0x02000000 0 0x30000000 0x30000000 0 0x2000
 				0x01000000 0 0x40000000 0x40000000 0 0x2000>;
-		pci at 8,0 {
-			compatible = "pci-generic";
-			reg = <0x4000 0 0 0 0>;
-			emul at 8,0 {
-				compatible = "sandbox,swap-case";
-			};
-		};
-		pci at c,0 {
-			compatible = "pci-generic";
-			reg = <0x6000 0 0 0 0>;
-			emul at c,0 {
-				compatible = "sandbox,swap-case";
-			};
-		};
+		sandbox,dev-info = <0x08 0x00 0x1234 0x5678
+				    0x0c 0x00 0x1234 0x5678>;
 	};
 
 	probing {
-- 
2.7.4

[U-Boot] [PATCH v2 16/21] test: dm: pci: Test driver binding with driver data provided

[Bin Meng]

With struct pci_device_id, it's possible to pass a driver data for
bound driver to use. This adds a test case for this functionality.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 test/dm/pci.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/test/dm/pci.c b/test/dm/pci.c
index 089b72e..53d2ca1 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -108,3 +108,20 @@ static int dm_test_pci_swapcase(struct unit_test_state *uts)
 	return 0;
 }
 DM_TEST(dm_test_pci_swapcase, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
+
+/* Test that we can dynamically bind the device driver correctly */
+static int dm_test_pci_drvdata(struct unit_test_state *uts)
+{
+	struct udevice *bus, *swap;
+
+	/* Check that asking for the device automatically fires up PCI */
+	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 1, &bus));
+
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x08, 0), &swap));
+	ut_asserteq(SWAP_CASE_DRV_DATA, swap->driver_data);
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x0c, 0), &swap));
+	ut_asserteq(SWAP_CASE_DRV_DATA, swap->driver_data);
+
+	return 0;
+}
+DM_TEST(dm_test_pci_drvdata, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
-- 
2.7.4

[U-Boot] [PATCH v2 17/21] pci: sandbox: emul: Rename priv structure

[Bin Meng]

We have "struct sandbox_pci_priv" in pci_sandbox driver. To avoid
confusion, rename the emul's priv to "struct sandbox_pci_emul_priv".

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 drivers/pci/pci-emul-uclass.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/pci-emul-uclass.c b/drivers/pci/pci-emul-uclass.c
index e9d2f49..3822758 100644
--- a/drivers/pci/pci-emul-uclass.c
+++ b/drivers/pci/pci-emul-uclass.c
@@ -11,7 +11,7 @@
 #include <pci.h>
 #include <dm/lists.h>
 
-struct sandbox_pci_priv {
+struct sandbox_pci_emul_priv {
 	int dev_count;
 };
 
@@ -43,7 +43,7 @@ int sandbox_pci_get_emul(struct udevice *bus, pci_dev_t find_devfn,
 
 static int sandbox_pci_emul_post_probe(struct udevice *dev)
 {
-	struct sandbox_pci_priv *priv = dev->uclass->priv;
+	struct sandbox_pci_emul_priv *priv = dev->uclass->priv;
 
 	priv->dev_count++;
 	sandbox_set_enable_pci_map(true);
@@ -53,7 +53,7 @@ static int sandbox_pci_emul_post_probe(struct udevice *dev)
 
 static int sandbox_pci_emul_pre_remove(struct udevice *dev)
 {
-	struct sandbox_pci_priv *priv = dev->uclass->priv;
+	struct sandbox_pci_emul_priv *priv = dev->uclass->priv;
 
 	priv->dev_count--;
 	sandbox_set_enable_pci_map(priv->dev_count > 0);
@@ -66,5 +66,5 @@ UCLASS_DRIVER(pci_emul) = {
 	.name		= "pci_emul",
 	.post_probe	= sandbox_pci_emul_post_probe,
 	.pre_remove	= sandbox_pci_emul_pre_remove,
-	.priv_auto_alloc_size	= sizeof(struct sandbox_pci_priv),
+	.priv_auto_alloc_size	= sizeof(struct sandbox_pci_emul_priv),
 };
-- 
2.7.4

[U-Boot] [PATCH v2 18/21] test: dm: pci: Add tests for mixed static and dynamic devices on the same bus

[Bin Meng]

In the Sandbox test configuration, PCI bus#0 only has static devices
while bus#1 only has dynamic devices. Create a bus#2 that has both
types of devices and test such.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

Changes in v2: None

 arch/sandbox/dts/test.dts | 18 ++++++++++++++
 test/dm/pci.c             | 63 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 81 insertions(+)

diff --git a/arch/sandbox/dts/test.dts b/arch/sandbox/dts/test.dts
index 44d24f9..7035646 100644
--- a/arch/sandbox/dts/test.dts
+++ b/arch/sandbox/dts/test.dts
@@ -16,6 +16,7 @@
 		mmc1 = "/mmc1";
 		pci0 = &pci0;
 		pci1 = &pci1;
+		pci2 = &pci2;
 		remoteproc1 = &rproc_1;
 		remoteproc2 = &rproc_2;
 		rtc0 = &rtc_0;
@@ -330,6 +331,23 @@
 				    0x0c 0x00 0x1234 0x5678>;
 	};
 
+	pci2: pci-controller2 {
+		compatible = "sandbox,pci";
+		device_type = "pci";
+		#address-cells = <3>;
+		#size-cells = <2>;
+		ranges = <0x02000000 0 0x50000000 0x50000000 0 0x2000
+				0x01000000 0 0x60000000 0x60000000 0 0x2000>;
+		sandbox,dev-info = <0x08 0x00 0x1234 0x5678>;
+		pci at 1f,0 {
+			compatible = "pci-generic";
+			reg = <0xf800 0 0 0 0>;
+			emul at 1f,0 {
+				compatible = "sandbox,swap-case";
+			};
+		};
+	};
+
 	probing {
 		compatible = "simple-bus";
 		test1 {
diff --git a/test/dm/pci.c b/test/dm/pci.c
index 53d2ca1..e03f6ba 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -125,3 +125,66 @@ static int dm_test_pci_drvdata(struct unit_test_state *uts)
 	return 0;
 }
 DM_TEST(dm_test_pci_drvdata, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
+
+/* Test that devices on PCI bus#2 can be accessed correctly */
+static int dm_test_pci_mixed(struct unit_test_state *uts)
+{
+	/* PCI bus#2 has both statically and dynamic declared devices */
+	struct udevice *bus, *swap;
+	u16 vendor, device;
+	ulong io_addr, mem_addr;
+	char *ptr;
+
+	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 2, &bus));
+
+	/* Test the dynamic device */
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(2, 0x08, 0), &swap));
+	vendor = 0;
+	ut_assertok(dm_pci_read_config16(swap, PCI_VENDOR_ID, &vendor));
+	ut_asserteq(SANDBOX_PCI_VENDOR_ID, vendor);
+
+	/* First test I/O */
+	io_addr = dm_pci_read_bar32(swap, 0);
+	outb(2, io_addr);
+	ut_asserteq(2, inb(io_addr));
+
+	/*
+	 * Now test memory mapping - note we must unmap and remap to cause
+	 * the swapcase emulation to see our data and response.
+	 */
+	mem_addr = dm_pci_read_bar32(swap, 1);
+	ptr = map_sysmem(mem_addr, 30);
+	strcpy(ptr, "This is a TesT oN dYNAMIc");
+	unmap_sysmem(ptr);
+
+	ptr = map_sysmem(mem_addr, 30);
+	ut_asserteq_str("tHIS IS A tESt On DynamiC", ptr);
+	unmap_sysmem(ptr);
+
+	/* Test the static device */
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(2, 0x1f, 0), &swap));
+	device = 0;
+	ut_assertok(dm_pci_read_config16(swap, PCI_DEVICE_ID, &device));
+	ut_asserteq(SANDBOX_PCI_DEVICE_ID, device);
+
+	/* First test I/O */
+	io_addr = dm_pci_read_bar32(swap, 0);
+	outb(2, io_addr);
+	ut_asserteq(2, inb(io_addr));
+
+	/*
+	 * Now test memory mapping - note we must unmap and remap to cause
+	 * the swapcase emulation to see our data and response.
+	 */
+	mem_addr = dm_pci_read_bar32(swap, 1);
+	ptr = map_sysmem(mem_addr, 30);
+	strcpy(ptr, "This is a TesT oN sTATIc");
+	unmap_sysmem(ptr);
+
+	ptr = map_sysmem(mem_addr, 30);
+	ut_asserteq_str("tHIS IS A tESt On StatiC", ptr);
+	unmap_sysmem(ptr);
+
+	return 0;
+}
+DM_TEST(dm_test_pci_mixed, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
-- 
2.7.4

[U-Boot] [PATCH v2 19/21] pci: Add all known capability and extended capability ids

[Bin Meng]

Currently we don't have a complete list of capability and extended
capability ids. This adds them.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
---

Changes in v2: None

 include/pci.h | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/include/pci.h b/include/pci.h
index c0ae5d1..83a40a5 100644
--- a/include/pci.h
+++ b/include/pci.h
@@ -318,7 +318,21 @@
 #define  PCI_CAP_ID_SLOTID	0x04	/* Slot Identification */
 #define  PCI_CAP_ID_MSI		0x05	/* Message Signalled Interrupts */
 #define  PCI_CAP_ID_CHSWP	0x06	/* CompactPCI HotSwap */
-#define  PCI_CAP_ID_EXP 	0x10	/* PCI Express */
+#define  PCI_CAP_ID_PCIX	0x07	/* PCI-X */
+#define  PCI_CAP_ID_HT		0x08	/* HyperTransport */
+#define  PCI_CAP_ID_VNDR	0x09	/* Vendor-Specific */
+#define  PCI_CAP_ID_DBG		0x0A	/* Debug port */
+#define  PCI_CAP_ID_CCRC	0x0B	/* CompactPCI Central Resource Control */
+#define  PCI_CAP_ID_SHPC	0x0C	/* PCI Standard Hot-Plug Controller */
+#define  PCI_CAP_ID_SSVID	0x0D	/* Bridge subsystem vendor/device ID */
+#define  PCI_CAP_ID_AGP3	0x0E	/* AGP Target PCI-PCI bridge */
+#define  PCI_CAP_ID_SECDEV	0x0F	/* Secure Device */
+#define  PCI_CAP_ID_EXP		0x10	/* PCI Express */
+#define  PCI_CAP_ID_MSIX	0x11	/* MSI-X */
+#define  PCI_CAP_ID_SATA	0x12	/* SATA Data/Index Conf. */
+#define  PCI_CAP_ID_AF		0x13	/* PCI Advanced Features */
+#define  PCI_CAP_ID_EA		0x14	/* PCI Enhanced Allocation */
+#define  PCI_CAP_ID_MAX		PCI_CAP_ID_EA
 #define PCI_CAP_LIST_NEXT	1	/* Next capability in the list */
 #define PCI_CAP_FLAGS		2	/* Capability defined flags (16 bits) */
 #define PCI_CAP_SIZEOF		4
@@ -434,6 +448,10 @@
 #define PCI_EXT_CAP_ID_SECPCI	0x19	/* Secondary PCIe Capability */
 #define PCI_EXT_CAP_ID_PMUX	0x1A	/* Protocol Multiplexing */
 #define PCI_EXT_CAP_ID_PASID	0x1B	/* Process Address Space ID */
+#define PCI_EXT_CAP_ID_DPC	0x1D	/* Downstream Port Containment */
+#define PCI_EXT_CAP_ID_L1SS	0x1E	/* L1 PM Substates */
+#define PCI_EXT_CAP_ID_PTM	0x1F	/* Precision Time Measurement */
+#define PCI_EXT_CAP_ID_MAX	PCI_EXT_CAP_ID_PTM
 
 /* Include the ID list */
 
-- 
2.7.4

[U-Boot] [PATCH v2 20/21] dm: pci: Add APIs to find capability and extended capability

[Bin Meng]

This introduces two new APIs dm_pci_find_capability() and
dm_pci_find_ext_capability() to get PCI capability address and
PCI express extended capability address for a given PCI device.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>

---

Changes in v2:
- Mask header_type with 0x7f

 drivers/pci/pci-uclass.c | 68 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/pci.h            | 46 ++++++++++++++++++++++++++++++++
 2 files changed, 114 insertions(+)

diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c
index 5eb6841..e9671d9 100644
--- a/drivers/pci/pci-uclass.c
+++ b/drivers/pci/pci-uclass.c
@@ -1320,6 +1320,74 @@ void *dm_pci_map_bar(struct udevice *dev, int bar, int flags)
 	return dm_pci_bus_to_virt(dev, pci_bus_addr, flags, 0, MAP_NOCACHE);
 }
 
+int dm_pci_find_capability(struct udevice *dev, int cap)
+{
+	u16 status;
+	u8 header_type;
+	int ttl = PCI_FIND_CAP_TTL;
+	u8 id;
+	u16 ent;
+	u8 pos;
+
+	dm_pci_read_config16(dev, PCI_STATUS, &status);
+	if (!(status & PCI_STATUS_CAP_LIST))
+		return 0;
+
+	dm_pci_read_config8(dev, PCI_HEADER_TYPE, &header_type);
+	if ((header_type & 0x7f) == PCI_HEADER_TYPE_CARDBUS)
+		pos = PCI_CB_CAPABILITY_LIST;
+	else
+		pos = PCI_CAPABILITY_LIST;
+
+	dm_pci_read_config8(dev, pos, &pos);
+	while (ttl--) {
+		if (pos < PCI_STD_HEADER_SIZEOF)
+			break;
+		pos &= ~3;
+		dm_pci_read_config16(dev, pos, &ent);
+
+		id = ent & 0xff;
+		if (id == 0xff)
+			break;
+		if (id == cap)
+			return pos;
+		pos = (ent >> 8);
+	}
+
+	return 0;
+}
+
+int dm_pci_find_ext_capability(struct udevice *dev, int cap)
+{
+	u32 header;
+	int ttl;
+	int pos = PCI_CFG_SPACE_SIZE;
+
+	/* minimum 8 bytes per capability */
+	ttl = (PCI_CFG_SPACE_EXP_SIZE - PCI_CFG_SPACE_SIZE) / 8;
+
+	dm_pci_read_config32(dev, pos, &header);
+	/*
+	 * If we have no capabilities, this is indicated by cap ID,
+	 * cap version and next pointer all being 0.
+	 */
+	if (header == 0)
+		return 0;
+
+	while (ttl--) {
+		if (PCI_EXT_CAP_ID(header) == cap)
+			return pos;
+
+		pos = PCI_EXT_CAP_NEXT(header);
+		if (pos < PCI_CFG_SPACE_SIZE)
+			break;
+
+		dm_pci_read_config32(dev, pos, &header);
+	}
+
+	return 0;
+}
+
 UCLASS_DRIVER(pci) = {
 	.id		= UCLASS_PCI,
 	.name		= "pci",
diff --git a/include/pci.h b/include/pci.h
index 83a40a5..938a839 100644
--- a/include/pci.h
+++ b/include/pci.h
@@ -17,6 +17,7 @@
  * Under PCI, each device has 256 bytes of configuration address space,
  * of which the first 64 bytes are standardized as follows:
  */
+#define PCI_STD_HEADER_SIZEOF	64
 #define PCI_VENDOR_ID		0x00	/* 16 bits */
 #define PCI_DEVICE_ID		0x02	/* 16 bits */
 #define PCI_COMMAND		0x04	/* 16 bits */
@@ -1311,6 +1312,51 @@ pci_addr_t dm_pci_phys_to_bus(struct udevice *dev, phys_addr_t addr,
  */
 void *dm_pci_map_bar(struct udevice *dev, int bar, int flags);
 
+/**
+ * dm_pci_find_capability() - find a capability
+ *
+ * Tell if a device supports a given PCI capability. Returns the
+ * address of the requested capability structure within the device's
+ * PCI configuration space or 0 in case the device does not support it.
+ *
+ * Possible values for @cap:
+ *
+ *  %PCI_CAP_ID_MSI	Message Signalled Interrupts
+ *  %PCI_CAP_ID_PCIX	PCI-X
+ *  %PCI_CAP_ID_EXP	PCI Express
+ *  %PCI_CAP_ID_MSIX	MSI-X
+ *
+ * See PCI_CAP_ID_xxx for the complete capability ID codes.
+ *
+ * @dev:	PCI device to query
+ * @cap:	capability code
+ * @return:	capability address or 0 if not supported
+ */
+int dm_pci_find_capability(struct udevice *dev, int cap);
+
+/**
+ * dm_pci_find_ext_capability() - find an extended capability
+ *
+ * Tell if a device supports a given PCI express extended capability.
+ * Returns the address of the requested extended capability structure
+ * within the device's PCI configuration space or 0 in case the device
+ * does not support it.
+ *
+ * Possible values for @cap:
+ *
+ *  %PCI_EXT_CAP_ID_ERR	Advanced Error Reporting
+ *  %PCI_EXT_CAP_ID_VC	Virtual Channel
+ *  %PCI_EXT_CAP_ID_DSN	Device Serial Number
+ *  %PCI_EXT_CAP_ID_PWR	Power Budgeting
+ *
+ * See PCI_EXT_CAP_ID_xxx for the complete extended capability ID codes.
+ *
+ * @dev:	PCI device to query
+ * @cap:	extended capability code
+ * @return:	extended capability address or 0 if not supported
+ */
+int dm_pci_find_ext_capability(struct udevice *dev, int cap);
+
 #define dm_pci_virt_to_bus(dev, addr, flags) \
 	dm_pci_phys_to_bus(dev, (virt_to_phys(addr)), (flags))
 #define dm_pci_bus_to_virt(dev, addr, flags, len, map_flags) \
-- 
2.7.4

[U-Boot] [PATCH v2 21/21] test: dm: pci: Add cases for finding PCI capability APIs

[Bin Meng]

Add several PCI capability and extended capability ID registers
in the swap_case driver, so that we can add test case for
dm_pci_find_capability() and dm_pci_find_ext_capability().

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>

---

Changes in v2: None

 arch/sandbox/include/asm/test.h |  8 ++++++++
 drivers/misc/swap_case.c        | 21 +++++++++++++++++++++
 test/dm/pci.c                   | 32 ++++++++++++++++++++++++++++++++
 3 files changed, 61 insertions(+)

diff --git a/arch/sandbox/include/asm/test.h b/arch/sandbox/include/asm/test.h
index 57aeca8..c8ae52b 100644
--- a/arch/sandbox/include/asm/test.h
+++ b/arch/sandbox/include/asm/test.h
@@ -16,6 +16,14 @@
 #define SANDBOX_PCI_CLASS_CODE		PCI_CLASS_CODE_COMM
 #define SANDBOX_PCI_CLASS_SUB_CODE	PCI_CLASS_SUB_CODE_COMM_SERIAL
 
+#define PCI_CAP_ID_PM_OFFSET		0x50
+#define PCI_CAP_ID_EXP_OFFSET		0x60
+#define PCI_CAP_ID_MSIX_OFFSET		0x70
+
+#define PCI_EXT_CAP_ID_ERR_OFFSET	0x100
+#define PCI_EXT_CAP_ID_VC_OFFSET	0x200
+#define PCI_EXT_CAP_ID_DSN_OFFSET	0x300
+
 /* Useful for PCI_VDEVICE() macro */
 #define PCI_VENDOR_ID_SANDBOX		SANDBOX_PCI_VENDOR_ID
 #define SWAP_CASE_DRV_DATA		0x55aa
diff --git a/drivers/misc/swap_case.c b/drivers/misc/swap_case.c
index 790bb0c..bffb809 100644
--- a/drivers/misc/swap_case.c
+++ b/drivers/misc/swap_case.c
@@ -118,6 +118,27 @@ static int sandbox_swap_case_read_config(struct udevice *emul, uint offset,
 		*valuep = result;
 		break;
 	}
+	case PCI_CAPABILITY_LIST:
+		*valuep = PCI_CAP_ID_PM_OFFSET;
+		break;
+	case PCI_CAP_ID_PM_OFFSET:
+		*valuep = (PCI_CAP_ID_EXP_OFFSET << 8) | PCI_CAP_ID_PM;
+		break;
+	case PCI_CAP_ID_EXP_OFFSET:
+		*valuep = (PCI_CAP_ID_MSIX_OFFSET << 8) | PCI_CAP_ID_EXP;
+		break;
+	case PCI_CAP_ID_MSIX_OFFSET:
+		*valuep = PCI_CAP_ID_MSIX;
+		break;
+	case PCI_EXT_CAP_ID_ERR_OFFSET:
+		*valuep = (PCI_EXT_CAP_ID_VC_OFFSET << 20) | PCI_EXT_CAP_ID_ERR;
+		break;
+	case PCI_EXT_CAP_ID_VC_OFFSET:
+		*valuep = (PCI_EXT_CAP_ID_DSN_OFFSET << 20) | PCI_EXT_CAP_ID_VC;
+		break;
+	case PCI_EXT_CAP_ID_DSN_OFFSET:
+		*valuep = PCI_EXT_CAP_ID_DSN;
+		break;
 	}
 
 	return 0;
diff --git a/test/dm/pci.c b/test/dm/pci.c
index e03f6ba..8699700 100644
--- a/test/dm/pci.c
+++ b/test/dm/pci.c
@@ -188,3 +188,35 @@ static int dm_test_pci_mixed(struct unit_test_state *uts)
 	return 0;
 }
 DM_TEST(dm_test_pci_mixed, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
+
+/* Test looking up PCI capability and extended capability */
+static int dm_test_pci_cap(struct unit_test_state *uts)
+{
+	struct udevice *bus, *swap;
+	int cap;
+
+	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 0, &bus));
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(0, 0x1f, 0), &swap));
+
+	/* look up PCI_CAP_ID_EXP */
+	cap = dm_pci_find_capability(swap, PCI_CAP_ID_EXP);
+	ut_asserteq(PCI_CAP_ID_EXP_OFFSET, cap);
+
+	/* look up PCI_CAP_ID_PCIX */
+	cap = dm_pci_find_capability(swap, PCI_CAP_ID_PCIX);
+	ut_asserteq(0, cap);
+
+	ut_assertok(uclass_get_device_by_seq(UCLASS_PCI, 1, &bus));
+	ut_assertok(dm_pci_bus_find_bdf(PCI_BDF(1, 0x08, 0), &swap));
+
+	/* look up PCI_EXT_CAP_ID_DSN */
+	cap = dm_pci_find_ext_capability(swap, PCI_EXT_CAP_ID_DSN);
+	ut_asserteq(PCI_EXT_CAP_ID_DSN_OFFSET, cap);
+
+	/* look up PCI_EXT_CAP_ID_SRIOV */
+	cap = dm_pci_find_ext_capability(swap, PCI_EXT_CAP_ID_SRIOV);
+	ut_asserteq(0, cap);
+
+	return 0;
+}
+DM_TEST(dm_test_pci_cap, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
-- 
2.7.4

[U-Boot] [PATCH v2 01/21] pci: Remove 440ep specific macros

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> These macros should not be put in the generic pci.h header file.
> Since they are not referenced anywhere, remove them completely.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  include/pci.h | 15 ---------------
>  1 file changed, 15 deletions(-)
>

Reviewed-by: Simon Glass <sjg@chromium.org>

nit for subject: 440ep-specific

[U-Boot] [PATCH v2 02/21] dm: Correct typos in uclass_first/next_device_check()

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> Correct typos in the comment block of uclass_first/next_device_check().
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  include/dm/uclass.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 03/21] dm: core: Add ofnode function to read PCI vendor and device id

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> We don't have the live-tree version of fdtdec_get_pci_vendev().
> This adds the API.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  drivers/core/ofnode.c | 36 ++++++++++++++++++++++++++++++++++++
>  include/dm/ofnode.h   | 13 +++++++++++++
>  2 files changed, 49 insertions(+)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 04/21] dm: pci: Extract vendor/device id in child_post_bind()

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> Currently only devfn is extracted in child_post_bind(). Now that
> we have the live-tree version API to look up PCI vendor and device
> id from the compatible string, let's extract and save them too.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  drivers/pci/pci-uclass.c | 11 +++++------
>  1 file changed, 5 insertions(+), 6 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 05/21] dm: pci: Fix scanning multi-function device

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> The flag to control whether to scan multi-function device during
> enumeration should be cleared at the beginning of each iteration
> if the device's function number equals to zero.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  drivers/pci/pci-uclass.c | 2 ++
>  1 file changed, 2 insertions(+)

Reviewed-by: Simon Glass <sjg@chromium.org>

Oops!

[U-Boot] [PATCH v2 06/21] test: dm: pci: Remove unnecessary steps in dm_test_pci_swapcase()

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> The check on uclass_get_device() and device_active() is unnecessary
> as the follow-up test operations will implicitly probe the driver.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  test/dm/pci.c | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 07/21] test: dm: pci: Test more than one device on the same bus

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> It's quite common to have more than one device on the same PCI bus.
> This updates the test case to test such scenario.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  arch/sandbox/dts/test.dts |  7 +++++++
>  test/dm/pci.c             | 37 +++++++++++++++++++++++++++++++++----
>  2 files changed, 40 insertions(+), 4 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 08/21] pci: sandbox: swap_case: Preserve space indicator bit in BAR registers

[Simon Glass]

On 29 July 2018 at 07:36, Bin Meng <bmeng.cn@gmail.com> wrote:
> With the newly added testing of more than one device, we get:
>
>   => ut dm pci_swapcase
>   Test: dm_test_pci_swapcase: pci.c
>   test/dm/pci.c:88, dm_test_pci_swapcase(): "tHIS IS A tESt" = ptr:
>   Expected "tHIS IS A tESt", got "this is a test"
>   Test: dm_test_pci_swapcase: pci.c (flat tree)
>   test/dm/pci.c:88, dm_test_pci_swapcase(): "tHIS IS A tESt" = ptr:
>   Expected "tHIS IS A tESt", got "this is a test"
>   Failures: 2
>
> The failure only happens on the 2nd swap_case device on the PCI bus.
> The case passes on the 1st device.
>
> It turns out the swap_case driver does not emulate bit#0 in BAR
> registers as a read-only bit. This corrects the implementation.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  drivers/misc/swap_case.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 09/21] test: dm: pci: Test more than one PCI host controller

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> So far there is only one PCI host controller in the sandbox test
> configuration. This is normally the case for x86, but it can be
> common on other architectures like ARM/PPC to have more than one
> PCI host controller in the system.
>
> This updates the case to cover such scenario.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  arch/sandbox/dts/test.dts | 28 ++++++++++++++++++++++++++--
>  test/dm/pci.c             | 11 +++++++++++
>  2 files changed, 37 insertions(+), 2 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 10/21] test: dm: pci: Add tests for configuration space access

[Simon Glass]

Hi Bin,

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> So far we missed the testing for PCI configuration space access.

You are also removing some asserts, but I suppose they are rendundant.

>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  test/dm/pci.c | 24 +++++++++++++++---------
>  1 file changed, 15 insertions(+), 9 deletions(-)
>

Reviewed-by: Simon Glass <sjg@chromium.org>

Regards,
Simon

[U-Boot] [PATCH v2 11/21] pci: sandbox: emul: Fix the call to pci_bus_find_devfn()

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> With the newly added test cases for PCI configuration access, we get:
>
>   => ut dm pci_busdev
>   Test: dm_test_pci_busdev: pci.c
>   test/dm/pci.c:49, dm_test_pci_busdev(): SANDBOX_PCI_VENDOR_ID == vendor:
>   Expected 4660, got 65535
>   Test: dm_test_pci_busdev: pci.c (flat tree)
>   test/dm/pci.c:49, dm_test_pci_busdev(): SANDBOX_PCI_VENDOR_ID == vendor:
>   Expected 4660, got 65535
>   Failures: 2
>
> The bug only shows up when bus number is not equal to zero. This is
> caused by the plain find_devfn parameter is passed to function call
> pci_bus_find_devfn(), inside which find_devfn is compared to devfn
> in the device's pplat structure. However pplat->devfn does not carry
> the bus number. Fix this by passing find_devfn with bus number masked.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  drivers/pci/pci-emul-uclass.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

Nice to find this.

[U-Boot] [PATCH v2 13/21] pci: sandbox: Support dynamically binding device driver

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> At present all emulated sandbox pci devices must be present in the
> device tree in order to be used. The real world pci uclass driver
> supports pci device driver matching, and we should add such support
> on sandbox too.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  doc/driver-model/pci-info.txt | 23 +++++++++++++
>  drivers/pci/pci-emul-uclass.c | 14 +++++---
>  drivers/pci/pci_sandbox.c     | 78 +++++++++++++++++++++++++++++++++++++++----
>  include/pci.h                 |  3 +-
>  4 files changed, 107 insertions(+), 11 deletions(-)
>

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 14/21] pci: sandbox: swap_case: Declare dynamic driver matching

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> This adds a U_BOOT_PCI_DEVICE() declaration to the swap_case driver.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  arch/sandbox/include/asm/test.h | 4 ++++
>  drivers/misc/swap_case.c        | 7 +++++++
>  2 files changed, 11 insertions(+)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 15/21] sandbox: Update test.dts for dynamic PCI device driver matching

[Simon Glass]

Hi Bin,

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  arch/sandbox/dts/test.dts | 16 ++--------------
>  1 file changed, 2 insertions(+), 14 deletions(-)

But doesn't this mean we lose testing of the device-tree config? We
need to keep that.

- Simon

[U-Boot] [PATCH v2 16/21] test: dm: pci: Test driver binding with driver data provided

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> With struct pci_device_id, it's possible to pass a driver data for
> bound driver to use. This adds a test case for this functionality.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  test/dm/pci.c | 17 +++++++++++++++++
>  1 file changed, 17 insertions(+)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 17/21] pci: sandbox: emul: Rename priv structure

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> We have "struct sandbox_pci_priv" in pci_sandbox driver. To avoid
> confusion, rename the emul's priv to "struct sandbox_pci_emul_priv".
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  drivers/pci/pci-emul-uclass.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 18/21] test: dm: pci: Add tests for mixed static and dynamic devices on the same bus

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> In the Sandbox test configuration, PCI bus#0 only has static devices
> while bus#1 only has dynamic devices. Create a bus#2 that has both
> types of devices and test such.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
> Changes in v2: None
>
>  arch/sandbox/dts/test.dts | 18 ++++++++++++++
>  test/dm/pci.c             | 63 +++++++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 81 insertions(+)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 20/21] dm: pci: Add APIs to find capability and extended capability

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> This introduces two new APIs dm_pci_find_capability() and
> dm_pci_find_ext_capability() to get PCI capability address and
> PCI express extended capability address for a given PCI device.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
>
> ---
>
> Changes in v2:
> - Mask header_type with 0x7f
>
>  drivers/pci/pci-uclass.c | 68 ++++++++++++++++++++++++++++++++++++++++++++++++
>  include/pci.h            | 46 ++++++++++++++++++++++++++++++++
>  2 files changed, 114 insertions(+)

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 21/21] test: dm: pci: Add cases for finding PCI capability APIs

[Simon Glass]

On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
> Add several PCI capability and extended capability ID registers
> in the swap_case driver, so that we can add test case for
> dm_pci_find_capability() and dm_pci_find_ext_capability().
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
>
> ---
>
> Changes in v2: None
>
>  arch/sandbox/include/asm/test.h |  8 ++++++++
>  drivers/misc/swap_case.c        | 21 +++++++++++++++++++++
>  test/dm/pci.c                   | 32 ++++++++++++++++++++++++++++++++
>  3 files changed, 61 insertions(+)
>

Reviewed-by: Simon Glass <sjg@chromium.org>

[U-Boot] [PATCH v2 10/21] test: dm: pci: Add tests for configuration space access

[Bin Meng]

Hi Simon,

On Fri, Aug 3, 2018 at 4:38 AM, Simon Glass <sjg@chromium.org> wrote:
> Hi Bin,
>
> On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
>> So far we missed the testing for PCI configuration space access.
>
> You are also removing some asserts, but I suppose they are rendundant.
>

Correct.

>>
>> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
>> ---
>>
>> Changes in v2: None
>>
>>  test/dm/pci.c | 24 +++++++++++++++---------
>>  1 file changed, 15 insertions(+), 9 deletions(-)
>>
>
> Reviewed-by: Simon Glass <sjg@chromium.org>
>

Regards,
Bin

[U-Boot] [PATCH v2 15/21] sandbox: Update test.dts for dynamic PCI device driver matching

[Bin Meng]

Hi Simon,

On Fri, Aug 3, 2018 at 4:38 AM, Simon Glass <sjg@chromium.org> wrote:
> Hi Bin,
>
> On 29 July 2018 at 07:37, Bin Meng <bmeng.cn@gmail.com> wrote:
>> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
>> ---
>>
>> Changes in v2: None
>>
>>  arch/sandbox/dts/test.dts | 16 ++--------------
>>  1 file changed, 2 insertions(+), 14 deletions(-)
>
> But doesn't this mean we lose testing of the device-tree config? We
> need to keep that.
>

The device-tree config testing is on bus 0. So there are 3 buses in
the test configuration.

bus 0: 2 device-tree devices
bus 1: 2 dynamic devices
bus 2: 1 device-tree and 1 dynamic (mixed config)

Regards,
Bin