[U-Boot] [PATCH v3 0/7] am57xx: cl-som-am57x: fix usb

[U-Boot] [PATCH v3 0/7] am57xx: cl-som-am57x: fix usb

[Uri Mashiach]

Various USB related comits for the CL-SOM-AM57x module.

---
V1 -> V2: Replace commit "fix XHCI registers base address" with "reintroduce the CONFIG_AM57XX symbol".
V2 -> V3: * New commit "move CONFIG_DRA7XX to Kconfig"
          * Replace commit "reintroduce the CONFIG_AM57XX symbol" with "xHCI registers based on USB port index"

Uri Mashiach (7):
  arm: dra7xx: move CONFIG_DRA7XX to Kconfig
  arm: usb: dra7xx: xHCI registers based on USB port index
  usb: host: xhci-omap: fix double weak board_usb_init functions
  arm: am57xx: cl-som-am57x: invoke clock API to enable/disable clocks
  arm: am57xx: cl-som-am57x: fix USB scan
  arm: am57xx: cl-som-am57x: enable USB storage
  arm: am57xx: cl-som-am57x: enable USB commands

 board/compulab/cl-som-am57x/Kconfig        |  4 ++++
 board/compulab/cl-som-am57x/cl-som-am57x.c | 10 ----------
 board/ti/am43xx/board.c                    |  4 ++--
 board/ti/am57xx/Kconfig                    |  4 ++++
 board/ti/am57xx/board.c                    |  4 ++--
 board/ti/dra7xx/Kconfig                    |  4 ++++
 board/ti/dra7xx/evm.c                      |  4 ++--
 configs/cl-som-am57x_defconfig             |  2 ++
 configs/dra7xx_evm_defconfig               |  1 +
 configs/dra7xx_hs_evm_defconfig            |  1 +
 drivers/usb/host/Kconfig                   | 16 ++++++++++++++++
 drivers/usb/host/xhci-omap.c               | 19 +++++++++++++++++--
 include/configs/am57xx_evm.h               |  2 --
 include/configs/cl-som-am57x.h             |  6 ++----
 include/configs/dra7xx_evm.h               |  2 --
 include/linux/usb/xhci-omap.h              |  6 ++++--
 scripts/config_whitelist.txt               |  1 -
 17 files changed, 61 insertions(+), 29 deletions(-)

-- 
2.7.4

[U-Boot] [PATCH v3 1/7] arm: dra7xx: move CONFIG_DRA7XX to Kconfig

[Uri Mashiach]

The symbol CONFIG_DRA7XX is needed for Kconfig conditions.

Cc: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
---
V3 -> V2: new commit

 board/compulab/cl-som-am57x/Kconfig | 4 ++++
 board/ti/am57xx/Kconfig             | 4 ++++
 board/ti/dra7xx/Kconfig             | 4 ++++
 include/configs/am57xx_evm.h        | 2 --
 include/configs/cl-som-am57x.h      | 2 --
 include/configs/dra7xx_evm.h        | 2 --
 scripts/config_whitelist.txt        | 1 -
 7 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/board/compulab/cl-som-am57x/Kconfig b/board/compulab/cl-som-am57x/Kconfig
index 85fc9a1..6d9ddc0 100644
--- a/board/compulab/cl-som-am57x/Kconfig
+++ b/board/compulab/cl-som-am57x/Kconfig
@@ -9,4 +9,8 @@ config SYS_VENDOR
 config SYS_CONFIG_NAME
 	default "cl-som-am57x"
 
+config DRA7XX
+	bool
+	default y
+
 endif
diff --git a/board/ti/am57xx/Kconfig b/board/ti/am57xx/Kconfig
index cead0f4..e70bcee 100644
--- a/board/ti/am57xx/Kconfig
+++ b/board/ti/am57xx/Kconfig
@@ -18,6 +18,10 @@ config CONS_INDEX
           Depending on your specific board you may want something other than UART3
           here.
 
+config DRA7XX
+	bool
+	default y
+
 source "board/ti/common/Kconfig"
 
 endif
diff --git a/board/ti/dra7xx/Kconfig b/board/ti/dra7xx/Kconfig
index b642113..bec3f1e 100644
--- a/board/ti/dra7xx/Kconfig
+++ b/board/ti/dra7xx/Kconfig
@@ -18,6 +18,10 @@ config CONS_INDEX
 	  Depending on your specific board you may want something other than UART1
 	  here.
 
+config DRA7XX
+	bool
+	default y
+
 source "board/ti/common/Kconfig"
 
 endif
diff --git a/include/configs/am57xx_evm.h b/include/configs/am57xx_evm.h
index 3d8b996..42cd434 100644
--- a/include/configs/am57xx_evm.h
+++ b/include/configs/am57xx_evm.h
@@ -14,8 +14,6 @@
 
 #include <environment/ti/dfu.h>
 
-#define CONFIG_DRA7XX
-
 #ifdef CONFIG_SPL_BUILD
 #define CONFIG_IODELAY_RECALIBRATION
 #endif
diff --git a/include/configs/cl-som-am57x.h b/include/configs/cl-som-am57x.h
index e1f724b..c5f39ff 100644
--- a/include/configs/cl-som-am57x.h
+++ b/include/configs/cl-som-am57x.h
@@ -11,8 +11,6 @@
 #ifndef __CONFIG_CL_SOM_AM57X_H
 #define __CONFIG_CL_SOM_AM57X_H
 
-#define CONFIG_DRA7XX
-
 #define CONFIG_NR_DRAM_BANKS		2
 
 #define CONSOLEDEV			"ttyO2"
diff --git a/include/configs/dra7xx_evm.h b/include/configs/dra7xx_evm.h
index bf8c041..3c7bdef 100644
--- a/include/configs/dra7xx_evm.h
+++ b/include/configs/dra7xx_evm.h
@@ -14,8 +14,6 @@
 
 #include <environment/ti/dfu.h>
 
-#define CONFIG_DRA7XX
-
 #ifdef CONFIG_SPL_BUILD
 #define CONFIG_IODELAY_RECALIBRATION
 #endif
diff --git a/scripts/config_whitelist.txt b/scripts/config_whitelist.txt
index d21589b..7f5b3c7 100644
--- a/scripts/config_whitelist.txt
+++ b/scripts/config_whitelist.txt
@@ -806,7 +806,6 @@ CONFIG_DNP5370_EXT_WD_DISABLE
 CONFIG_DP_DDR_CTRL
 CONFIG_DP_DDR_DIMM_SLOTS_PER_CTLR
 CONFIG_DP_DDR_NUM_CTRLS
-CONFIG_DRA7XX
 CONFIG_DRAM_2G
 CONFIG_DRAM_TIMINGS_
 CONFIG_DRIVER_AT91EMAC
-- 
2.7.4

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Uri Mashiach]

Modify the determination of the base address of xHCI registers of DRA7XX
targets.
Before the commit: by the target.
After the commit: by the USB port index.

Cc: Lokesh Vutla <lokeshvutla@ti.com>
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
---
V1 -> V2: Replace the commit "fix XHCI registers base address".
V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"

 configs/dra7xx_evm_defconfig    |  1 +
 configs/dra7xx_hs_evm_defconfig |  1 +
 drivers/usb/host/Kconfig        | 16 ++++++++++++++++
 include/linux/usb/xhci-omap.h   |  6 ++++--
 4 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
index 26b26cc..1f47f92 100644
--- a/configs/dra7xx_evm_defconfig
+++ b/configs/dra7xx_evm_defconfig
@@ -86,6 +86,7 @@ CONFIG_OMAP_TIMER=y
 CONFIG_USB=y
 CONFIG_USB_XHCI_HCD=y
 CONFIG_USB_XHCI_DWC3=y
+CONFIG_USB_XHCI_DRA7XX_INDEX1=y
 CONFIG_USB_DWC3=y
 CONFIG_USB_DWC3_GADGET=y
 CONFIG_USB_DWC3_OMAP=y
diff --git a/configs/dra7xx_hs_evm_defconfig b/configs/dra7xx_hs_evm_defconfig
index 244940c..5addd3e 100644
--- a/configs/dra7xx_hs_evm_defconfig
+++ b/configs/dra7xx_hs_evm_defconfig
@@ -91,6 +91,7 @@ CONFIG_OMAP_TIMER=y
 CONFIG_USB=y
 CONFIG_USB_XHCI_HCD=y
 CONFIG_USB_XHCI_DWC3=y
+CONFIG_USB_XHCI_DRA7XX_INDEX1=y
 CONFIG_USB_DWC3=y
 CONFIG_USB_DWC3_GADGET=y
 CONFIG_USB_DWC3_OMAP=y
diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index 5129a57..440fbcf 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
 	help
 	  Enables support for the on-chip xHCI controller on Xilinx ZynqMP SoCs.
 
+choice
+	prompt "DRA7XX xHCI USB index select"
+	depends on DRA7XX
+
+config USB_XHCI_DRA7XX_INDEX0
+	bool "USB0"
+	help
+	  DRA7XX xHCI USB0.
+
+config USB_XHCI_DRA7XX_INDEX1
+	bool "USB1"
+	help
+	  DRA7XX xHCI USB1.
+
+endchoice
+
 endif # USB_XHCI_HCD
 
 config USB_EHCI_HCD
diff --git a/include/linux/usb/xhci-omap.h b/include/linux/usb/xhci-omap.h
index 9de80d7..3f65a76 100644
--- a/include/linux/usb/xhci-omap.h
+++ b/include/linux/usb/xhci-omap.h
@@ -10,14 +10,16 @@
 #ifndef _ASM_ARCH_XHCI_OMAP_H_
 #define _ASM_ARCH_XHCI_OMAP_H_
 
-#ifdef CONFIG_TARGET_DRA7XX_EVM
+#ifdef CONFIG_DRA7XX
+#ifdef CONFIG_USB_XHCI_DRA7XX_INDEX1
 #define OMAP_XHCI_BASE 0x488d0000
 #define OMAP_OCP1_SCP_BASE 0x4A081000
 #define OMAP_OTG_WRAPPER_BASE 0x488c0000
-#elif defined CONFIG_TARGET_AM57XX_EVM
+#elif defined CONFIG_USB_XHCI_DRA7XX_INDEX0
 #define OMAP_XHCI_BASE 0x48890000
 #define OMAP_OCP1_SCP_BASE 0x4A084c00
 #define OMAP_OTG_WRAPPER_BASE 0x48880000
+#endif /* CONFIG_USB_XHCI_DRA7XX_INDEX1 */
 #elif defined CONFIG_AM43XX
 #define OMAP_XHCI_BASE 0x483d0000
 #define OMAP_OCP1_SCP_BASE 0x483E8000
-- 
2.7.4

[U-Boot] [PATCH v3 3/7] usb: host: xhci-omap: fix double weak board_usb_init functions

[Uri Mashiach]

A weak version of the function board_usb_init is implemented in:
common/usb.c
drivers/usb/host/xhci-omap.c

To fix the double implementations:
* Convert the board_usb_init function in drivers/usb/host/xhci-omap.c
  normal (not weak).
* The function board_usb_init in drivers/usb/host/xhci-omap.c calls to
  the weak function omap_xhci_board_usb_init.
* Rename board version of the function board_usb_init to
  omap_xhci_board_usb_init.
  Done only for boards that defines CONFIG_USB_XHCI_OMAP.

To achieve the same flexibility with the function board_usb_cleanup:
* Add a normal (not weak) implementation of the function
  board_usb_cleanup in drivers/usb/host/xhci-omap.c
* The function board_usb_cleanup in drivers/usb/host/xhci-omap.c calls
  to the weak function omap_xhci_board_usb_cleanup.
* Rename board version of the function board_usb_cleanup to
  omap_xhci_board_usb_cleanup.
  Done only for boards that defines CONFIG_USB_XHCI_OMAP.

Cc: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Acked-by: Marek Vasut <marex@denx.de>
Reviewed-by: Tom Rini <trini@konsulko.com>
---
V1 -> V2: Use __weak instead of attribute block
V2 -> V3: none

 board/compulab/cl-som-am57x/cl-som-am57x.c |  2 +-
 board/ti/am43xx/board.c                    |  4 ++--
 board/ti/am57xx/board.c                    |  4 ++--
 board/ti/dra7xx/evm.c                      |  4 ++--
 drivers/usb/host/xhci-omap.c               | 17 +++++++++++++++--
 5 files changed, 22 insertions(+), 9 deletions(-)

diff --git a/board/compulab/cl-som-am57x/cl-som-am57x.c b/board/compulab/cl-som-am57x/cl-som-am57x.c
index bdd0a2b..fe1468f 100644
--- a/board/compulab/cl-som-am57x/cl-som-am57x.c
+++ b/board/compulab/cl-som-am57x/cl-som-am57x.c
@@ -54,7 +54,7 @@ int board_mmc_init(bd_t *bis)
 #endif /* CONFIG_GENERIC_MMC */
 
 #ifdef CONFIG_USB_XHCI_OMAP
-int board_usb_init(int index, enum usb_init_type init)
+int omap_xhci_board_usb_init(int index, enum usb_init_type init)
 {
 	setbits_le32((*prcm)->cm_l3init_usb_otg_ss1_clkctrl,
 		     OTG_SS_CLKCTRL_MODULEMODE_HW | OPTFCLKEN_REFCLK960M);
diff --git a/board/ti/am43xx/board.c b/board/ti/am43xx/board.c
index 390cc16..2572029 100644
--- a/board/ti/am43xx/board.c
+++ b/board/ti/am43xx/board.c
@@ -694,7 +694,7 @@ int usb_gadget_handle_interrupts(int index)
 #endif /* CONFIG_USB_DWC3 */
 
 #if defined(CONFIG_USB_DWC3) || defined(CONFIG_USB_XHCI_OMAP)
-int board_usb_init(int index, enum usb_init_type init)
+int omap_xhci_board_usb_init(int index, enum usb_init_type init)
 {
 	enable_usb_clocks(index);
 #ifdef CONFIG_USB_DWC3
@@ -725,7 +725,7 @@ int board_usb_init(int index, enum usb_init_type init)
 	return 0;
 }
 
-int board_usb_cleanup(int index, enum usb_init_type init)
+int omap_xhci_board_usb_cleanup(int index, enum usb_init_type init)
 {
 #ifdef CONFIG_USB_DWC3
 	switch (index) {
diff --git a/board/ti/am57xx/board.c b/board/ti/am57xx/board.c
index 5f2d4df..96fad0f 100644
--- a/board/ti/am57xx/board.c
+++ b/board/ti/am57xx/board.c
@@ -616,7 +616,7 @@ int usb_gadget_handle_interrupts(int index)
 #endif /* CONFIG_USB_DWC3 */
 
 #if defined(CONFIG_USB_DWC3) || defined(CONFIG_USB_XHCI_OMAP)
-int board_usb_init(int index, enum usb_init_type init)
+int omap_xhci_board_usb_init(int index, enum usb_init_type init)
 {
 	enable_usb_clocks(index);
 	switch (index) {
@@ -650,7 +650,7 @@ int board_usb_init(int index, enum usb_init_type init)
 	return 0;
 }
 
-int board_usb_cleanup(int index, enum usb_init_type init)
+int omap_xhci_board_usb_cleanup(int index, enum usb_init_type init)
 {
 #ifdef CONFIG_USB_DWC3
 	switch (index) {
diff --git a/board/ti/dra7xx/evm.c b/board/ti/dra7xx/evm.c
index bd1c809..65bce93 100644
--- a/board/ti/dra7xx/evm.c
+++ b/board/ti/dra7xx/evm.c
@@ -727,7 +727,7 @@ static struct ti_usb_phy_device usb_phy2_device = {
 	.index = 1,
 };
 
-int board_usb_init(int index, enum usb_init_type init)
+int omap_xhci_board_usb_init(int index, enum usb_init_type init)
 {
 	enable_usb_clocks(index);
 	switch (index) {
@@ -764,7 +764,7 @@ int board_usb_init(int index, enum usb_init_type init)
 	return 0;
 }
 
-int board_usb_cleanup(int index, enum usb_init_type init)
+int omap_xhci_board_usb_cleanup(int index, enum usb_init_type init)
 {
 	switch (index) {
 	case 0:
diff --git a/drivers/usb/host/xhci-omap.c b/drivers/usb/host/xhci-omap.c
index b881b19..a1b4f2f 100644
--- a/drivers/usb/host/xhci-omap.c
+++ b/drivers/usb/host/xhci-omap.c
@@ -27,12 +27,25 @@ DECLARE_GLOBAL_DATA_PTR;
 
 static struct omap_xhci omap;
 
-__weak int __board_usb_init(int index, enum usb_init_type init)
+__weak int omap_xhci_board_usb_init(int index, enum usb_init_type init)
 {
 	return 0;
 }
+
 int board_usb_init(int index, enum usb_init_type init)
-	__attribute__((weak, alias("__board_usb_init")));
+{
+	return omap_xhci_board_usb_init(index, init);
+}
+
+__weak int omap_xhci_board_usb_cleanup(int index, enum usb_init_type init)
+{
+	return 0;
+}
+
+int board_usb_cleanup(int index, enum usb_init_type init)
+{
+	return omap_xhci_board_usb_cleanup(index, init);
+}
 
 static int omap_xhci_core_init(struct omap_xhci *omap)
 {
-- 
2.7.4

[U-Boot] [PATCH v3 4/7] arm: am57xx: cl-som-am57x: invoke clock API to enable/disable clocks

[Uri Mashiach]

Invoke enable_usb_clocks during board_usb_init and disable_usb_clocks
during board_usb_exit to enable and disable clocks respectively.

Modifications:
* Enable USB clocks in the OMAP version of the function
  board_usb_init.
* Disable USB clocks in the OMAP version of the function
  board_usb_cleanup.

Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Reviewed-by: Marek Vasut <marex@denx.de>
Reviewed-by: Tom Rini <trini@konsulko.com>
---
V1 -> V3: none

 board/compulab/cl-som-am57x/cl-som-am57x.c | 10 ----------
 drivers/usb/host/xhci-omap.c               |  2 ++
 2 files changed, 2 insertions(+), 10 deletions(-)

diff --git a/board/compulab/cl-som-am57x/cl-som-am57x.c b/board/compulab/cl-som-am57x/cl-som-am57x.c
index fe1468f..4701b71 100644
--- a/board/compulab/cl-som-am57x/cl-som-am57x.c
+++ b/board/compulab/cl-som-am57x/cl-som-am57x.c
@@ -53,16 +53,6 @@ int board_mmc_init(bd_t *bis)
 }
 #endif /* CONFIG_GENERIC_MMC */
 
-#ifdef CONFIG_USB_XHCI_OMAP
-int omap_xhci_board_usb_init(int index, enum usb_init_type init)
-{
-	setbits_le32((*prcm)->cm_l3init_usb_otg_ss1_clkctrl,
-		     OTG_SS_CLKCTRL_MODULEMODE_HW | OPTFCLKEN_REFCLK960M);
-
-	return 0;
-}
-#endif /* CONFIG_USB_XHCI_OMAP */
-
 int misc_init_r(void)
 {
 	cl_print_pcb_info();
diff --git a/drivers/usb/host/xhci-omap.c b/drivers/usb/host/xhci-omap.c
index a1b4f2f..d6c5744 100644
--- a/drivers/usb/host/xhci-omap.c
+++ b/drivers/usb/host/xhci-omap.c
@@ -29,6 +29,7 @@ static struct omap_xhci omap;
 
 __weak int omap_xhci_board_usb_init(int index, enum usb_init_type init)
 {
+	enable_usb_clocks(index);
 	return 0;
 }
 
@@ -39,6 +40,7 @@ int board_usb_init(int index, enum usb_init_type init)
 
 __weak int omap_xhci_board_usb_cleanup(int index, enum usb_init_type init)
 {
+	disable_usb_clocks(index);
 	return 0;
 }
 
-- 
2.7.4

[U-Boot] [PATCH v3 5/7] arm: am57xx: cl-som-am57x: fix USB scan

[Uri Mashiach]

USB bus scan attempt:
----------------------------------cut----------------------------------
=> usb start
starting USB...
USB0:   Register 2000140 NbrPorts 2
Starting the controller
USB XHCI 1.00
scanning bus 0 for devices... data abort
pc : [<fff6240e>]          lr : [<fff623b3>]
reloc pc : [<8081b40e>]    lr : [<8081b3b3>]
sp : fdf42930  ip : fdf42960     fp : 00000000
r10: 00000001  r9 : fdf42ef0     r8 : 48890020
r7 : 00000002  r6 : fffa5840     r5 : fff8b140  r4 : fdf429c0
r3 : 00000000  r2 : 00000004     r1 : 00000000  r0 : 00000000
Flags: nZcv  IRQs off  FIQs off  Mode SVC_32
Resetting CPU ...

resetting ...
----------------------------------cut----------------------------------

Fix by enabling USB configuration in the SPL.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Igor Grinberg <grinberg@compulab.co.il>
---
V1 -> V3: none

 include/configs/cl-som-am57x.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/configs/cl-som-am57x.h b/include/configs/cl-som-am57x.h
index c5f39ff..3418005 100644
--- a/include/configs/cl-som-am57x.h
+++ b/include/configs/cl-som-am57x.h
@@ -83,6 +83,8 @@
 #define CONFIG_SYS_I2C_PCA953X_ADDR     0x20
 #define CONFIG_SYS_I2C_PCA953X_WIDTH    { {0x20, 16} }
 
+#endif /* !CONFIG_SPL_BUILD */
+
 /* USB xHCI HOST */
 #define CONFIG_USB_XHCI_OMAP
 #define CONFIG_SYS_USB_XHCI_MAX_ROOT_PORTS 2
@@ -97,8 +99,6 @@
 #define CONFIG_USB_ETHER_ASIX
 #define CONFIG_USB_ETHER_MCS7830
 
-#endif /* !CONFIG_SPL_BUILD */
-
 /* CPSW Ethernet */
 #define CONFIG_DRIVER_TI_CPSW
 #define CONFIG_MII
-- 
2.7.4

[U-Boot] [PATCH v3 6/7] arm: am57xx: cl-som-am57x: enable USB storage

[Uri Mashiach]

Add CONFIG_USB_STORAGE to the defconfig file.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Reviewed-by: Tom Rini <trini@konsulko.com>
---
V1 -> V3: none

 configs/cl-som-am57x_defconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/configs/cl-som-am57x_defconfig b/configs/cl-som-am57x_defconfig
index 916b836..1831f45 100644
--- a/configs/cl-som-am57x_defconfig
+++ b/configs/cl-som-am57x_defconfig
@@ -47,4 +47,5 @@ CONFIG_TI_QSPI=y
 CONFIG_USB=y
 CONFIG_USB_XHCI_HCD=y
 CONFIG_USB_XHCI_DWC3=y
+CONFIG_USB_STORAGE=y
 CONFIG_OF_LIBFDT=y
-- 
2.7.4

[U-Boot] [PATCH v3 7/7] arm: am57xx: cl-som-am57x: enable USB commands

[Uri Mashiach]

Add CONFIG_CMD_USB to the defconfig file.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Reviewed-by: Tom Rini <trini@konsulko.com>
---
V1 -> V3: none

 configs/cl-som-am57x_defconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/configs/cl-som-am57x_defconfig b/configs/cl-som-am57x_defconfig
index 1831f45..83d7bd0 100644
--- a/configs/cl-som-am57x_defconfig
+++ b/configs/cl-som-am57x_defconfig
@@ -15,6 +15,7 @@ CONFIG_CMD_PART=y
 CONFIG_CMD_SF=y
 CONFIG_CMD_SPI=y
 CONFIG_CMD_I2C=y
+CONFIG_CMD_USB=y
 CONFIG_CMD_GPIO=y
 # CONFIG_CMD_SETEXPR is not set
 CONFIG_CMD_DHCP=y
-- 
2.7.4

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Marek Vasut]

On 02/19/2017 02:27 PM, Uri Mashiach wrote:
> Modify the determination of the base address of xHCI registers of DRA7XX
> targets.
> Before the commit: by the target.
> After the commit: by the USB port index.
> 
> Cc: Lokesh Vutla <lokeshvutla@ti.com>
> Cc: Marek Vasut <marex@denx.de>
> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
> ---
> V1 -> V2: Replace the commit "fix XHCI registers base address".
> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
> 
>  configs/dra7xx_evm_defconfig    |  1 +
>  configs/dra7xx_hs_evm_defconfig |  1 +
>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>  include/linux/usb/xhci-omap.h   |  6 ++++--
>  4 files changed, 22 insertions(+), 2 deletions(-)
> 
> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
> index 26b26cc..1f47f92 100644
> --- a/configs/dra7xx_evm_defconfig
> +++ b/configs/dra7xx_evm_defconfig
> @@ -86,6 +86,7 @@ CONFIG_OMAP_TIMER=y
>  CONFIG_USB=y
>  CONFIG_USB_XHCI_HCD=y
>  CONFIG_USB_XHCI_DWC3=y
> +CONFIG_USB_XHCI_DRA7XX_INDEX1=y
>  CONFIG_USB_DWC3=y
>  CONFIG_USB_DWC3_GADGET=y
>  CONFIG_USB_DWC3_OMAP=y
> diff --git a/configs/dra7xx_hs_evm_defconfig b/configs/dra7xx_hs_evm_defconfig
> index 244940c..5addd3e 100644
> --- a/configs/dra7xx_hs_evm_defconfig
> +++ b/configs/dra7xx_hs_evm_defconfig
> @@ -91,6 +91,7 @@ CONFIG_OMAP_TIMER=y
>  CONFIG_USB=y
>  CONFIG_USB_XHCI_HCD=y
>  CONFIG_USB_XHCI_DWC3=y
> +CONFIG_USB_XHCI_DRA7XX_INDEX1=y
>  CONFIG_USB_DWC3=y
>  CONFIG_USB_DWC3_GADGET=y
>  CONFIG_USB_DWC3_OMAP=y
> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
> index 5129a57..440fbcf 100644
> --- a/drivers/usb/host/Kconfig
> +++ b/drivers/usb/host/Kconfig
> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>  	help
>  	  Enables support for the on-chip xHCI controller on Xilinx ZynqMP SoCs.
>  
> +choice
> +	prompt "DRA7XX xHCI USB index select"
> +	depends on DRA7XX
> +
> +config USB_XHCI_DRA7XX_INDEX0
> +	bool "USB0"
> +	help
> +	  DRA7XX xHCI USB0.
> +
> +config USB_XHCI_DRA7XX_INDEX1
> +	bool "USB1"
> +	help
> +	  DRA7XX xHCI USB1.

What is this all about ? Shouldn't this come from DT ? And what if I
want to use both XHCI ? This looks totally bogus ...

> +endchoice
> +
>  endif # USB_XHCI_HCD
>  
>  config USB_EHCI_HCD
> diff --git a/include/linux/usb/xhci-omap.h b/include/linux/usb/xhci-omap.h
> index 9de80d7..3f65a76 100644
> --- a/include/linux/usb/xhci-omap.h
> +++ b/include/linux/usb/xhci-omap.h
> @@ -10,14 +10,16 @@
>  #ifndef _ASM_ARCH_XHCI_OMAP_H_
>  #define _ASM_ARCH_XHCI_OMAP_H_
>  
> -#ifdef CONFIG_TARGET_DRA7XX_EVM
> +#ifdef CONFIG_DRA7XX
> +#ifdef CONFIG_USB_XHCI_DRA7XX_INDEX1
>  #define OMAP_XHCI_BASE 0x488d0000
>  #define OMAP_OCP1_SCP_BASE 0x4A081000
>  #define OMAP_OTG_WRAPPER_BASE 0x488c0000
> -#elif defined CONFIG_TARGET_AM57XX_EVM
> +#elif defined CONFIG_USB_XHCI_DRA7XX_INDEX0
>  #define OMAP_XHCI_BASE 0x48890000
>  #define OMAP_OCP1_SCP_BASE 0x4A084c00
>  #define OMAP_OTG_WRAPPER_BASE 0x48880000
> +#endif /* CONFIG_USB_XHCI_DRA7XX_INDEX1 */
>  #elif defined CONFIG_AM43XX
>  #define OMAP_XHCI_BASE 0x483d0000
>  #define OMAP_OCP1_SCP_BASE 0x483E8000
> 


-- 
Best regards,
Marek Vasut

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Uri Mashiach]

On 02/19/2017 04:27 PM, Marek Vasut wrote:
> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>> Modify the determination of the base address of xHCI registers of DRA7XX
>> targets.
>> Before the commit: by the target.
>> After the commit: by the USB port index.
>>
>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>> Cc: Marek Vasut <marex@denx.de>
>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>> ---
>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>
>>  configs/dra7xx_evm_defconfig    |  1 +
>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>
>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>> index 26b26cc..1f47f92 100644

[...]

>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>> index 5129a57..440fbcf 100644
>> --- a/drivers/usb/host/Kconfig
>> +++ b/drivers/usb/host/Kconfig
>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>  	help
>>  	  Enables support for the on-chip xHCI controller on Xilinx ZynqMP SoCs.
>>
>> +choice
>> +	prompt "DRA7XX xHCI USB index select"
>> +	depends on DRA7XX
>> +
>> +config USB_XHCI_DRA7XX_INDEX0
>> +	bool "USB0"
>> +	help
>> +	  DRA7XX xHCI USB0.
>> +
>> +config USB_XHCI_DRA7XX_INDEX1
>> +	bool "USB1"
>> +	help
>> +	  DRA7XX xHCI USB1.
>
> What is this all about ? Shouldn't this come from DT ? And what if I
> want to use both XHCI ? This looks totally bogus ...
>

The support for both XHCI is currently missing.
This could be a temporary solution until the DT solution.
The current situation is worse - selecting USB0 or USB1 based on the target.

>> +endchoice
>> +
>>  endif # USB_XHCI_HCD
>>
>>  config USB_EHCI_HCD
>> diff --git a/include/linux/usb/xhci-omap.h b/include/linux/usb/xhci-omap.h
>> index 9de80d7..3f65a76 100644
>> --- a/include/linux/usb/xhci-omap.h
>> +++ b/include/linux/usb/xhci-omap.h
>> @@ -10,14 +10,16 @@
>>  #ifndef _ASM_ARCH_XHCI_OMAP_H_

[...]

>>  #define OMAP_OCP1_SCP_BASE 0x483E8000
>>
>
>

-- 
Regards,
Uri

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Marek Vasut]

On 02/19/2017 03:55 PM, Uri Mashiach wrote:
> 
> 
> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>> targets.
>>> Before the commit: by the target.
>>> After the commit: by the USB port index.
>>>
>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>> Cc: Marek Vasut <marex@denx.de>
>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>> ---
>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>
>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>> index 26b26cc..1f47f92 100644
> 
> [...]
> 
>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>> index 5129a57..440fbcf 100644
>>> --- a/drivers/usb/host/Kconfig
>>> +++ b/drivers/usb/host/Kconfig
>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>      help
>>>        Enables support for the on-chip xHCI controller on Xilinx
>>> ZynqMP SoCs.
>>>
>>> +choice
>>> +    prompt "DRA7XX xHCI USB index select"
>>> +    depends on DRA7XX
>>> +
>>> +config USB_XHCI_DRA7XX_INDEX0
>>> +    bool "USB0"
>>> +    help
>>> +      DRA7XX xHCI USB0.
>>> +
>>> +config USB_XHCI_DRA7XX_INDEX1
>>> +    bool "USB1"
>>> +    help
>>> +      DRA7XX xHCI USB1.
>>
>> What is this all about ? Shouldn't this come from DT ? And what if I
>> want to use both XHCI ? This looks totally bogus ...
>>
> 
> The support for both XHCI is currently missing.
> This could be a temporary solution until the DT solution.
> The current situation is worse - selecting USB0 or USB1 based on the
> target.

So we're replacing it with equally bad solution ? Hmmm , no.
The MW will open mid-march, there's about a month to fix this,
so please do.


-- 
Best regards,
Marek Vasut

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Tom Rini]

On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
> > 
> > 
> > On 02/19/2017 04:27 PM, Marek Vasut wrote:
> >> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
> >>> Modify the determination of the base address of xHCI registers of DRA7XX
> >>> targets.
> >>> Before the commit: by the target.
> >>> After the commit: by the USB port index.
> >>>
> >>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
> >>> Cc: Marek Vasut <marex@denx.de>
> >>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
> >>> ---
> >>> V1 -> V2: Replace the commit "fix XHCI registers base address".
> >>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
> >>>
> >>>  configs/dra7xx_evm_defconfig    |  1 +
> >>>  configs/dra7xx_hs_evm_defconfig |  1 +
> >>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
> >>>  include/linux/usb/xhci-omap.h   |  6 ++++--
> >>>  4 files changed, 22 insertions(+), 2 deletions(-)
> >>>
> >>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
> >>> index 26b26cc..1f47f92 100644
> > 
> > [...]
> > 
> >>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
> >>> index 5129a57..440fbcf 100644
> >>> --- a/drivers/usb/host/Kconfig
> >>> +++ b/drivers/usb/host/Kconfig
> >>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
> >>>      help
> >>>        Enables support for the on-chip xHCI controller on Xilinx
> >>> ZynqMP SoCs.
> >>>
> >>> +choice
> >>> +    prompt "DRA7XX xHCI USB index select"
> >>> +    depends on DRA7XX
> >>> +
> >>> +config USB_XHCI_DRA7XX_INDEX0
> >>> +    bool "USB0"
> >>> +    help
> >>> +      DRA7XX xHCI USB0.
> >>> +
> >>> +config USB_XHCI_DRA7XX_INDEX1
> >>> +    bool "USB1"
> >>> +    help
> >>> +      DRA7XX xHCI USB1.
> >>
> >> What is this all about ? Shouldn't this come from DT ? And what if I
> >> want to use both XHCI ? This looks totally bogus ...
> >>
> > 
> > The support for both XHCI is currently missing.
> > This could be a temporary solution until the DT solution.
> > The current situation is worse - selecting USB0 or USB1 based on the
> > target.
> 
> So we're replacing it with equally bad solution ? Hmmm , no.
> The MW will open mid-march, there's about a month to fix this,
> so please do.

Do note that the relevant driver here is not yet using DM_USB.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170219/60625185/attachment.sig>

[U-Boot] [PATCH v3 1/7] arm: dra7xx: move CONFIG_DRA7XX to Kconfig

[Tom Rini]

On Sun, Feb 19, 2017 at 03:27:13PM +0200, Uri Mashiach wrote:

> The symbol CONFIG_DRA7XX is needed for Kconfig conditions.
> 
> Cc: Lokesh Vutla <lokeshvutla@ti.com>
> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
> ---
> V3 -> V2: new commit
> 
>  board/compulab/cl-som-am57x/Kconfig | 4 ++++
>  board/ti/am57xx/Kconfig             | 4 ++++
>  board/ti/dra7xx/Kconfig             | 4 ++++
>  include/configs/am57xx_evm.h        | 2 --
>  include/configs/cl-som-am57x.h      | 2 --
>  include/configs/dra7xx_evm.h        | 2 --
>  scripts/config_whitelist.txt        | 1 -
>  7 files changed, 12 insertions(+), 7 deletions(-)

We need to introduce the symbol in arch/arm/mach-omap2/omap5/Kconfig and
then select it under the appropriate boards.  Thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170219/f628e838/attachment.sig>

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Marek Vasut]

On 02/19/2017 04:15 PM, Tom Rini wrote:
> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>
>>>
>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>> targets.
>>>>> Before the commit: by the target.
>>>>> After the commit: by the USB port index.
>>>>>
>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>> ---
>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>
>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>
>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>> index 26b26cc..1f47f92 100644
>>>
>>> [...]
>>>
>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>> index 5129a57..440fbcf 100644
>>>>> --- a/drivers/usb/host/Kconfig
>>>>> +++ b/drivers/usb/host/Kconfig
>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>      help
>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>> ZynqMP SoCs.
>>>>>
>>>>> +choice
>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>> +    depends on DRA7XX
>>>>> +
>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>> +    bool "USB0"
>>>>> +    help
>>>>> +      DRA7XX xHCI USB0.
>>>>> +
>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>> +    bool "USB1"
>>>>> +    help
>>>>> +      DRA7XX xHCI USB1.
>>>>
>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>> want to use both XHCI ? This looks totally bogus ...
>>>>
>>>
>>> The support for both XHCI is currently missing.
>>> This could be a temporary solution until the DT solution.
>>> The current situation is worse - selecting USB0 or USB1 based on the
>>> target.
>>
>> So we're replacing it with equally bad solution ? Hmmm , no.
>> The MW will open mid-march, there's about a month to fix this,
>> so please do.
> 
> Do note that the relevant driver here is not yet using DM_USB.
> 
We had multi-controller support before DM existed, no problem.
But then, xhci does support DM, so fixing this should also not
be a problem.

-- 
Best regards,
Marek Vasut

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Igor Grinberg]

Hi guys,

On 02/19/17 17:15, Tom Rini wrote:
> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>
>>>
>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>> targets.
>>>>> Before the commit: by the target.
>>>>> After the commit: by the USB port index.
>>>>>
>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>> ---
>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>
>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>
>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>> index 26b26cc..1f47f92 100644
>>>
>>> [...]
>>>
>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>> index 5129a57..440fbcf 100644
>>>>> --- a/drivers/usb/host/Kconfig
>>>>> +++ b/drivers/usb/host/Kconfig
>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>      help
>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>> ZynqMP SoCs.
>>>>>
>>>>> +choice
>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>> +    depends on DRA7XX
>>>>> +
>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>> +    bool "USB0"
>>>>> +    help
>>>>> +      DRA7XX xHCI USB0.
>>>>> +
>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>> +    bool "USB1"
>>>>> +    help
>>>>> +      DRA7XX xHCI USB1.
>>>>
>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>> want to use both XHCI ? This looks totally bogus ...

Right, both XHCIs cannot be used with current driver and we do not have
the time to fix it by our own... may be TI has?
Remember, you've accepted the driver and following patches, right?

Regarding DT, do we have a DT as a requirement to run USB in U-Boot?
I don't remember this happening and I think it shouldn't be a requirement.

>>>>
>>>
>>> The support for both XHCI is currently missing.
>>> This could be a temporary solution until the DT solution.
>>> The current situation is worse - selecting USB0 or USB1 based on the
>>> target.
>>
>> So we're replacing it with equally bad solution ?

I don't think equally applies here...
This IS an improvement. Of course it is not like you would want it to be,
but still it is from a platform POV.

>> Hmmm , no.
>> The MW will open mid-march, there's about a month to fix this,
>> so please do.
> 
> Do note that the relevant driver here is not yet using DM_USB.

Yes, the driver should be fixed some day. We would really like to take
this task, but unfortunately, we cannot, at least not right now.
But we do need that USB working on our board and not only on TI EVMs...

Tom,
Should we fall back to v1 and have a worse solution for the base addresses?

-- 
Regards,
Igor.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 837 bytes
Desc: OpenPGP digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170219/be8c5535/attachment.sig>

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Igor Grinberg]

On 02/19/17 17:52, Marek Vasut wrote:
> On 02/19/2017 04:15 PM, Tom Rini wrote:
>> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>>
>>>>
>>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>>> targets.
>>>>>> Before the commit: by the target.
>>>>>> After the commit: by the USB port index.
>>>>>>
>>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>>> ---
>>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>>
>>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>>
>>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>>> index 26b26cc..1f47f92 100644
>>>>
>>>> [...]
>>>>
>>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>>> index 5129a57..440fbcf 100644
>>>>>> --- a/drivers/usb/host/Kconfig
>>>>>> +++ b/drivers/usb/host/Kconfig
>>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>>      help
>>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>>> ZynqMP SoCs.
>>>>>>
>>>>>> +choice
>>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>>> +    depends on DRA7XX
>>>>>> +
>>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>>> +    bool "USB0"
>>>>>> +    help
>>>>>> +      DRA7XX xHCI USB0.
>>>>>> +
>>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>>> +    bool "USB1"
>>>>>> +    help
>>>>>> +      DRA7XX xHCI USB1.
>>>>>
>>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>>> want to use both XHCI ? This looks totally bogus ...
>>>>>
>>>>
>>>> The support for both XHCI is currently missing.
>>>> This could be a temporary solution until the DT solution.
>>>> The current situation is worse - selecting USB0 or USB1 based on the
>>>> target.
>>>
>>> So we're replacing it with equally bad solution ? Hmmm , no.
>>> The MW will open mid-march, there's about a month to fix this,
>>> so please do.
>>
>> Do note that the relevant driver here is not yet using DM_USB.
>>
> We had multi-controller support before DM existed, no problem.
> But then, xhci does support DM, so fixing this should also not
> be a problem.

Right! We're talking on fixing the XHCI driver...
We can't afford it right now...

-- 
Regards,
Igor.

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Marek Vasut]

On 02/19/2017 05:26 PM, Igor Grinberg wrote:
> Hi guys,
> 
> On 02/19/17 17:15, Tom Rini wrote:
>> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>>
>>>>
>>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>>> targets.
>>>>>> Before the commit: by the target.
>>>>>> After the commit: by the USB port index.
>>>>>>
>>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>>> ---
>>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>>
>>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>>
>>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>>> index 26b26cc..1f47f92 100644
>>>>
>>>> [...]
>>>>
>>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>>> index 5129a57..440fbcf 100644
>>>>>> --- a/drivers/usb/host/Kconfig
>>>>>> +++ b/drivers/usb/host/Kconfig
>>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>>      help
>>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>>> ZynqMP SoCs.
>>>>>>
>>>>>> +choice
>>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>>> +    depends on DRA7XX
>>>>>> +
>>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>>> +    bool "USB0"
>>>>>> +    help
>>>>>> +      DRA7XX xHCI USB0.
>>>>>> +
>>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>>> +    bool "USB1"
>>>>>> +    help
>>>>>> +      DRA7XX xHCI USB1.
>>>>>
>>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>>> want to use both XHCI ? This looks totally bogus ...
> 
> Right, both XHCIs cannot be used with current driver and we do not have
> the time to fix it by our own... may be TI has?
> Remember, you've accepted the driver and following patches, right?

You mean this patch [1] which actually adds the board dependency ?
Actually , no , I DID NOT. It's been going on for a while that my
role as a USB custodian was actively circumvented and patches which
should've gone through the USB tree just did not.

So please, next time you start blaming someone, get your facts right.

I already have almost zero motivation to maintain USB in my free time,
being paid nothing for it, not ever hearing a single "thank you" and
just spending time I could use otherwise. Being circumvented and only
catching shit for problematic patches I did NOT even apply though, that
is completely off-putting .

[1] https://patchwork.ozlabs.org/patch/700372/

> Regarding DT, do we have a DT as a requirement to run USB in U-Boot?

It is highly recommended.

> I don't remember this happening and I think it shouldn't be a requirement.
> 
>>>>>
>>>>
>>>> The support for both XHCI is currently missing.
>>>> This could be a temporary solution until the DT solution.
>>>> The current situation is worse - selecting USB0 or USB1 based on the
>>>> target.
>>>
>>> So we're replacing it with equally bad solution ?
> 
> I don't think equally applies here...
> This IS an improvement. Of course it is not like you would want it to be,
> but still it is from a platform POV.

TBH, revering 042fdb7cabb8270eb86c45f11263fa91c12e3b65 might be the way
to go.

>>> Hmmm , no.
>>> The MW will open mid-march, there's about a month to fix this,
>>> so please do.
>>
>> Do note that the relevant driver here is not yet using DM_USB.
> 
> Yes, the driver should be fixed some day. We would really like to take
> this task, but unfortunately, we cannot, at least not right now.
> But we do need that USB working on our board and not only on TI EVMs...

See above, I believe 042fdb7cabb8270eb86c45f11263fa91c12e3b65 is bogus.

> Tom,
> Should we fall back to v1 and have a worse solution for the base addresses?

OK, I see me being USB custodian has exactly zero value now ... good to
know.

-- 
Best regards,
Marek Vasut

[U-Boot] [PATCH v3 1/7] arm: dra7xx: move CONFIG_DRA7XX to Kconfig

[Uri Mashiach]

On 02/19/2017 05:15 PM, Tom Rini wrote:
> On Sun, Feb 19, 2017 at 03:27:13PM +0200, Uri Mashiach wrote:
>
>> The symbol CONFIG_DRA7XX is needed for Kconfig conditions.
>>
>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>> ---
>> V3 -> V2: new commit
>>
>>  board/compulab/cl-som-am57x/Kconfig | 4 ++++
>>  board/ti/am57xx/Kconfig             | 4 ++++
>>  board/ti/dra7xx/Kconfig             | 4 ++++
>>  include/configs/am57xx_evm.h        | 2 --
>>  include/configs/cl-som-am57x.h      | 2 --
>>  include/configs/dra7xx_evm.h        | 2 --
>>  scripts/config_whitelist.txt        | 1 -
>>  7 files changed, 12 insertions(+), 7 deletions(-)
>
> We need to introduce the symbol in arch/arm/mach-omap2/omap5/Kconfig and
> then select it under the appropriate boards.  Thanks!
>

Maybe instead of selecting it under the appropriate board it can be 
defaulted to yes in the board's Kconfig?

-- 
Thanks and regards,
Uri

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Igor Grinberg]

Hey Marek,

We haven't spoken for great while... I'm glad to hear from you!

On 02/19/17 19:39, Marek Vasut wrote:
> On 02/19/2017 05:26 PM, Igor Grinberg wrote:
>> Hi guys,
>>
>> On 02/19/17 17:15, Tom Rini wrote:
>>> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>>>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>>>
>>>>>
>>>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>>>> targets.
>>>>>>> Before the commit: by the target.
>>>>>>> After the commit: by the USB port index.
>>>>>>>
>>>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>>>> ---
>>>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>>>
>>>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>>>
>>>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>>>> index 26b26cc..1f47f92 100644
>>>>>
>>>>> [...]
>>>>>
>>>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>>>> index 5129a57..440fbcf 100644
>>>>>>> --- a/drivers/usb/host/Kconfig
>>>>>>> +++ b/drivers/usb/host/Kconfig
>>>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>>>      help
>>>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>>>> ZynqMP SoCs.
>>>>>>>
>>>>>>> +choice
>>>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>>>> +    depends on DRA7XX
>>>>>>> +
>>>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>>>> +    bool "USB0"
>>>>>>> +    help
>>>>>>> +      DRA7XX xHCI USB0.
>>>>>>> +
>>>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>>>> +    bool "USB1"
>>>>>>> +    help
>>>>>>> +      DRA7XX xHCI USB1.
>>>>>>
>>>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>>>> want to use both XHCI ? This looks totally bogus ...
>>
>> Right, both XHCIs cannot be used with current driver and we do not have
>> the time to fix it by our own... may be TI has?
>> Remember, you've accepted the driver and following patches, right?
> 
> You mean this patch [1] which actually adds the board dependency ?
> Actually , no , I DID NOT. It's been going on for a while that my
> role as a USB custodian was actively circumvented and patches which
> should've gone through the USB tree just did not.

Ahh.. Sorry, thought you are collecting all the USB patches or at least
provide your Ack so patches can go through other trees..

> 
> So please, next time you start blaming someone, get your facts right.

Yeah, I haven't looked at it thoroughly... That's why the "right?" in the
end of my sentence. Sorry once again, I did not mean to offend you.

> 
> I already have almost zero motivation to maintain USB in my free time,
> being paid nothing for it, not ever hearing a single "thank you" and
> just spending time I could use otherwise. Being circumvented and only
> catching shit for problematic patches I did NOT even apply though, that
> is completely off-putting .

Ahhhhh... First of all thank you Marek! You've done a great job AFAIR.
I really mean it.
And it is a pity, things have gone this way.
May be this can still be fixed? Tom?

> 
> [1] https://patchwork.ozlabs.org/patch/700372/
> 
>> Regarding DT, do we have a DT as a requirement to run USB in U-Boot?
> 
> It is highly recommended.

Hmmm... I don't really like this approach as it does not help during the board
production, where I'd like to see the bootloader detect as much a it can
during runtime and then flash the board with correct DT...

> 
>> I don't remember this happening and I think it shouldn't be a requirement.
>>
>>>>>>
>>>>>
>>>>> The support for both XHCI is currently missing.
>>>>> This could be a temporary solution until the DT solution.
>>>>> The current situation is worse - selecting USB0 or USB1 based on the
>>>>> target.
>>>>
>>>> So we're replacing it with equally bad solution ?
>>
>> I don't think equally applies here...
>> This IS an improvement. Of course it is not like you would want it to be,
>> but still it is from a platform POV.
> 
> TBH, revering 042fdb7cabb8270eb86c45f11263fa91c12e3b65 might be the way
> to go.

Ok. I'm fine with this approach.
I'm also fine with if the driver gets fixed and both XHCI instances can be used.

> 
>>>> Hmmm , no.
>>>> The MW will open mid-march, there's about a month to fix this,
>>>> so please do.
>>>
>>> Do note that the relevant driver here is not yet using DM_USB.
>>
>> Yes, the driver should be fixed some day. We would really like to take
>> this task, but unfortunately, we cannot, at least not right now.
>> But we do need that USB working on our board and not only on TI EVMs...
> 
> See above, I believe 042fdb7cabb8270eb86c45f11263fa91c12e3b65 is bogus.

Well, agreed.

> 
>> Tom,
>> Should we fall back to v1 and have a worse solution for the base addresses?
> 
> OK, I see me being USB custodian has exactly zero value now ... good to
> know.

I believe we should start fixing things and probably the best time
would be... now? I mean, we can't fix things in the past, right?

-- 
Regards,
Igor.

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Marek Vasut]

On 02/20/2017 08:47 AM, Igor Grinberg wrote:
> Hey Marek,

Hi,

> We haven't spoken for great while... I'm glad to hear from you!

yeah

> On 02/19/17 19:39, Marek Vasut wrote:
>> On 02/19/2017 05:26 PM, Igor Grinberg wrote:
>>> Hi guys,
>>>
>>> On 02/19/17 17:15, Tom Rini wrote:
>>>> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>>>>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>>>>
>>>>>>
>>>>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>>>>> targets.
>>>>>>>> Before the commit: by the target.
>>>>>>>> After the commit: by the USB port index.
>>>>>>>>
>>>>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>>>>> ---
>>>>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>>>>
>>>>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>>>>
>>>>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>>>>> index 26b26cc..1f47f92 100644
>>>>>>
>>>>>> [...]
>>>>>>
>>>>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>>>>> index 5129a57..440fbcf 100644
>>>>>>>> --- a/drivers/usb/host/Kconfig
>>>>>>>> +++ b/drivers/usb/host/Kconfig
>>>>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>>>>      help
>>>>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>>>>> ZynqMP SoCs.
>>>>>>>>
>>>>>>>> +choice
>>>>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>>>>> +    depends on DRA7XX
>>>>>>>> +
>>>>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>>>>> +    bool "USB0"
>>>>>>>> +    help
>>>>>>>> +      DRA7XX xHCI USB0.
>>>>>>>> +
>>>>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>>>>> +    bool "USB1"
>>>>>>>> +    help
>>>>>>>> +      DRA7XX xHCI USB1.
>>>>>>>
>>>>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>>>>> want to use both XHCI ? This looks totally bogus ...
>>>
>>> Right, both XHCIs cannot be used with current driver and we do not have
>>> the time to fix it by our own... may be TI has?
>>> Remember, you've accepted the driver and following patches, right?
>>
>> You mean this patch [1] which actually adds the board dependency ?
>> Actually , no , I DID NOT. It's been going on for a while that my
>> role as a USB custodian was actively circumvented and patches which
>> should've gone through the USB tree just did not.
> 
> Ahh.. Sorry, thought you are collecting all the USB patches or at least
> provide your Ack so patches can go through other trees..

So did I ...

>> So please, next time you start blaming someone, get your facts right.
> 
> Yeah, I haven't looked at it thoroughly... That's why the "right?" in the
> end of my sentence. Sorry once again, I did not mean to offend you.
> 
>>
>> I already have almost zero motivation to maintain USB in my free time,
>> being paid nothing for it, not ever hearing a single "thank you" and
>> just spending time I could use otherwise. Being circumvented and only
>> catching shit for problematic patches I did NOT even apply though, that
>> is completely off-putting .
> 
> Ahhhhh... First of all thank you Marek! You've done a great job AFAIR.
> I really mean it.
> And it is a pity, things have gone this way.
> May be this can still be fixed? Tom?

I already had a long discussion with Tom yesterday ... so let's see .

>> [1] https://patchwork.ozlabs.org/patch/700372/
>>
>>> Regarding DT, do we have a DT as a requirement to run USB in U-Boot?
>>
>> It is highly recommended.
> 
> Hmmm... I don't really like this approach as it does not help during the board
> production, where I'd like to see the bootloader detect as much a it can
> during runtime and then flash the board with correct DT...

You can enable all during production, plug USB sticks into every port,
check which controllers detect devices and then only keep those enabled
via the "status" DT prop. As a bonus, you get functional test.

Does the OMAP5 not use DT yet ?

>>> I don't remember this happening and I think it shouldn't be a requirement.
>>>
>>>>>>>
>>>>>>
>>>>>> The support for both XHCI is currently missing.
>>>>>> This could be a temporary solution until the DT solution.
>>>>>> The current situation is worse - selecting USB0 or USB1 based on the
>>>>>> target.
>>>>>
>>>>> So we're replacing it with equally bad solution ?
>>>
>>> I don't think equally applies here...
>>> This IS an improvement. Of course it is not like you would want it to be,
>>> but still it is from a platform POV.
>>
>> TBH, revering 042fdb7cabb8270eb86c45f11263fa91c12e3b65 might be the way
>> to go.
> 
> Ok. I'm fine with this approach.
> I'm also fine with if the driver gets fixed and both XHCI instances can be used.

Someone would have to put that effort in.

I wonder if instead of defining INDEX0, INDEX1 ... this could be turned
into integer ...

>>
>>>>> Hmmm , no.
>>>>> The MW will open mid-march, there's about a month to fix this,
>>>>> so please do.
>>>>
>>>> Do note that the relevant driver here is not yet using DM_USB.
>>>
>>> Yes, the driver should be fixed some day. We would really like to take
>>> this task, but unfortunately, we cannot, at least not right now.
>>> But we do need that USB working on our board and not only on TI EVMs...
>>
>> See above, I believe 042fdb7cabb8270eb86c45f11263fa91c12e3b65 is bogus.
> 
> Well, agreed.
> 
>>
>>> Tom,
>>> Should we fall back to v1 and have a worse solution for the base addresses?
>>
>> OK, I see me being USB custodian has exactly zero value now ... good to
>> know.
> 
> I believe we should start fixing things and probably the best time
> would be... now? I mean, we can't fix things in the past, right?
> 


-- 
Best regards,
Marek Vasut

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Roger Quadros]

Hi,

On 19/02/17 18:26, Igor Grinberg wrote:
> Hi guys,
> 
> On 02/19/17 17:15, Tom Rini wrote:
>> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>>
>>>>
>>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>>> targets.
>>>>>> Before the commit: by the target.
>>>>>> After the commit: by the USB port index.
>>>>>>
>>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>>> ---
>>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>>
>>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>>
>>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>>> index 26b26cc..1f47f92 100644
>>>>
>>>> [...]
>>>>
>>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>>> index 5129a57..440fbcf 100644
>>>>>> --- a/drivers/usb/host/Kconfig
>>>>>> +++ b/drivers/usb/host/Kconfig
>>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>>      help
>>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>>> ZynqMP SoCs.
>>>>>>
>>>>>> +choice
>>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>>> +    depends on DRA7XX
>>>>>> +
>>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>>> +    bool "USB0"
>>>>>> +    help
>>>>>> +      DRA7XX xHCI USB0.
>>>>>> +
>>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>>> +    bool "USB1"
>>>>>> +    help
>>>>>> +      DRA7XX xHCI USB1.
>>>>>
>>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>>> want to use both XHCI ? This looks totally bogus ...
> 
> Right, both XHCIs cannot be used with current driver and we do not have
> the time to fix it by our own... may be TI has?

This is still a work in progress for us
http://lists.denx.de/pipermail/u-boot/2016-January/239931.html

> Remember, you've accepted the driver and following patches, right?
> 
> Regarding DT, do we have a DT as a requirement to run USB in U-Boot?
> I don't remember this happening and I think it shouldn't be a requirement.
> 
>>>>>
>>>>
>>>> The support for both XHCI is currently missing.
>>>> This could be a temporary solution until the DT solution.
>>>> The current situation is worse - selecting USB0 or USB1 based on the
>>>> target.
>>>
>>> So we're replacing it with equally bad solution ?
> 
> I don't think equally applies here...
> This IS an improvement. Of course it is not like you would want it to be,
> but still it is from a platform POV.
> 
>>> Hmmm , no.
>>> The MW will open mid-march, there's about a month to fix this,
>>> so please do.
>>
>> Do note that the relevant driver here is not yet using DM_USB.
> 
> Yes, the driver should be fixed some day. We would really like to take
> this task, but unfortunately, we cannot, at least not right now.
> But we do need that USB working on our board and not only on TI EVMs...
> 
> Tom,
> Should we fall back to v1 and have a worse solution for the base addresses?
> 

-- 
cheers,
-roger

[U-Boot] [PATCH v3 1/7] arm: dra7xx: move CONFIG_DRA7XX to Kconfig

[Tom Rini]

On Mon, Feb 20, 2017 at 09:18:47AM +0200, Uri Mashiach wrote:
> 
> 
> On 02/19/2017 05:15 PM, Tom Rini wrote:
> >On Sun, Feb 19, 2017 at 03:27:13PM +0200, Uri Mashiach wrote:
> >
> >>The symbol CONFIG_DRA7XX is needed for Kconfig conditions.
> >>
> >>Cc: Lokesh Vutla <lokeshvutla@ti.com>
> >>Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
> >>---
> >>V3 -> V2: new commit
> >>
> >> board/compulab/cl-som-am57x/Kconfig | 4 ++++
> >> board/ti/am57xx/Kconfig             | 4 ++++
> >> board/ti/dra7xx/Kconfig             | 4 ++++
> >> include/configs/am57xx_evm.h        | 2 --
> >> include/configs/cl-som-am57x.h      | 2 --
> >> include/configs/dra7xx_evm.h        | 2 --
> >> scripts/config_whitelist.txt        | 1 -
> >> 7 files changed, 12 insertions(+), 7 deletions(-)
> >
> >We need to introduce the symbol in arch/arm/mach-omap2/omap5/Kconfig and
> >then select it under the appropriate boards.  Thanks!
> 
> Maybe instead of selecting it under the appropriate board it can be
> defaulted to yes in the board's Kconfig?

No.  The correct thing is to select it (if it's not something that can
be opted out of, for example what type of SoC this is) or imply'd if
it's something that really should be enabled, but the user can opt out
of.  Doing a default yes in the board Kconfig was a stop-gap until
'imply' came around.  Thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170220/47575020/attachment.sig>

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Uri Mashiach]

On 02/20/2017 10:28 AM, Marek Vasut wrote:
> On 02/20/2017 08:47 AM, Igor Grinberg wrote:
>> Hey Marek,
>
> Hi,
>
>> We haven't spoken for great while... I'm glad to hear from you!
>
> yeah
>
>> On 02/19/17 19:39, Marek Vasut wrote:
>>> On 02/19/2017 05:26 PM, Igor Grinberg wrote:
>>>> Hi guys,
>>>>
>>>> On 02/19/17 17:15, Tom Rini wrote:
>>>>> On Sun, Feb 19, 2017 at 04:13:02PM +0100, Marek Vasut wrote:
>>>>>> On 02/19/2017 03:55 PM, Uri Mashiach wrote:
>>>>>>>
>>>>>>>
>>>>>>> On 02/19/2017 04:27 PM, Marek Vasut wrote:
>>>>>>>> On 02/19/2017 02:27 PM, Uri Mashiach wrote:
>>>>>>>>> Modify the determination of the base address of xHCI registers of DRA7XX
>>>>>>>>> targets.
>>>>>>>>> Before the commit: by the target.
>>>>>>>>> After the commit: by the USB port index.
>>>>>>>>>
>>>>>>>>> Cc: Lokesh Vutla <lokeshvutla@ti.com>
>>>>>>>>> Cc: Marek Vasut <marex@denx.de>
>>>>>>>>> Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
>>>>>>>>> ---
>>>>>>>>> V1 -> V2: Replace the commit "fix XHCI registers base address".
>>>>>>>>> V2 -> V3: Replace the commit "reintroduce the CONFIG_AM57XX symbol"
>>>>>>>>>
>>>>>>>>>  configs/dra7xx_evm_defconfig    |  1 +
>>>>>>>>>  configs/dra7xx_hs_evm_defconfig |  1 +
>>>>>>>>>  drivers/usb/host/Kconfig        | 16 ++++++++++++++++
>>>>>>>>>  include/linux/usb/xhci-omap.h   |  6 ++++--
>>>>>>>>>  4 files changed, 22 insertions(+), 2 deletions(-)
>>>>>>>>>
>>>>>>>>> diff --git a/configs/dra7xx_evm_defconfig b/configs/dra7xx_evm_defconfig
>>>>>>>>> index 26b26cc..1f47f92 100644
>>>>>>>
>>>>>>> [...]
>>>>>>>
>>>>>>>>> diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
>>>>>>>>> index 5129a57..440fbcf 100644
>>>>>>>>> --- a/drivers/usb/host/Kconfig
>>>>>>>>> +++ b/drivers/usb/host/Kconfig
>>>>>>>>> @@ -43,6 +43,22 @@ config USB_XHCI_ZYNQMP
>>>>>>>>>      help
>>>>>>>>>        Enables support for the on-chip xHCI controller on Xilinx
>>>>>>>>> ZynqMP SoCs.
>>>>>>>>>
>>>>>>>>> +choice
>>>>>>>>> +    prompt "DRA7XX xHCI USB index select"
>>>>>>>>> +    depends on DRA7XX
>>>>>>>>> +
>>>>>>>>> +config USB_XHCI_DRA7XX_INDEX0
>>>>>>>>> +    bool "USB0"
>>>>>>>>> +    help
>>>>>>>>> +      DRA7XX xHCI USB0.
>>>>>>>>> +
>>>>>>>>> +config USB_XHCI_DRA7XX_INDEX1
>>>>>>>>> +    bool "USB1"
>>>>>>>>> +    help
>>>>>>>>> +      DRA7XX xHCI USB1.
>>>>>>>>
>>>>>>>> What is this all about ? Shouldn't this come from DT ? And what if I
>>>>>>>> want to use both XHCI ? This looks totally bogus ...
>>>>
>>>> Right, both XHCIs cannot be used with current driver and we do not have
>>>> the time to fix it by our own... may be TI has?
>>>> Remember, you've accepted the driver and following patches, right?
>>>
>>> You mean this patch [1] which actually adds the board dependency ?
>>> Actually , no , I DID NOT. It's been going on for a while that my
>>> role as a USB custodian was actively circumvented and patches which
>>> should've gone through the USB tree just did not.
>>
>> Ahh.. Sorry, thought you are collecting all the USB patches or at least
>> provide your Ack so patches can go through other trees..
>
> So did I ...
>
>>> So please, next time you start blaming someone, get your facts right.
>>
>> Yeah, I haven't looked at it thoroughly... That's why the "right?" in the
>> end of my sentence. Sorry once again, I did not mean to offend you.
>>
>>>
>>> I already have almost zero motivation to maintain USB in my free time,
>>> being paid nothing for it, not ever hearing a single "thank you" and
>>> just spending time I could use otherwise. Being circumvented and only
>>> catching shit for problematic patches I did NOT even apply though, that
>>> is completely off-putting .
>>
>> Ahhhhh... First of all thank you Marek! You've done a great job AFAIR.
>> I really mean it.
>> And it is a pity, things have gone this way.
>> May be this can still be fixed? Tom?
>
> I already had a long discussion with Tom yesterday ... so let's see .
>
>>> [1] https://patchwork.ozlabs.org/patch/700372/
>>>
>>>> Regarding DT, do we have a DT as a requirement to run USB in U-Boot?
>>>
>>> It is highly recommended.
>>
>> Hmmm... I don't really like this approach as it does not help during the board
>> production, where I'd like to see the bootloader detect as much a it can
>> during runtime and then flash the board with correct DT...
>
> You can enable all during production, plug USB sticks into every port,
> check which controllers detect devices and then only keep those enabled
> via the "status" DT prop. As a bonus, you get functional test.
>
> Does the OMAP5 not use DT yet ?
>
>>>> I don't remember this happening and I think it shouldn't be a requirement.
>>>>
>>>>>>>>
>>>>>>>
>>>>>>> The support for both XHCI is currently missing.
>>>>>>> This could be a temporary solution until the DT solution.
>>>>>>> The current situation is worse - selecting USB0 or USB1 based on the
>>>>>>> target.
>>>>>>
>>>>>> So we're replacing it with equally bad solution ?
>>>>
>>>> I don't think equally applies here...
>>>> This IS an improvement. Of course it is not like you would want it to be,
>>>> but still it is from a platform POV.
>>>
>>> TBH, revering 042fdb7cabb8270eb86c45f11263fa91c12e3b65 might be the way
>>> to go.
>>
>> Ok. I'm fine with this approach.
>> I'm also fine with if the driver gets fixed and both XHCI instances can be used.
>
> Someone would have to put that effort in.
>
> I wonder if instead of defining INDEX0, INDEX1 ... this could be turned
> into integer ...
>

Something like the following?

#ifdef CONFIG_DRA7XX
#if CONFIG_USB_XHCI_DRA7XX_INDEX == 1
...
#elif CONFIG_USB_XHCI_DRA7XX_INDEX == 0
...
#endif /* CONFIG_USB_XHCI_DRA7XX_INDEX == 1 */
#elif defined CONFIG_AM43XX
...

>>>
>>>>>> Hmmm , no.
>>>>>> The MW will open mid-march, there's about a month to fix this,
>>>>>> so please do.
>>>>>
>>>>> Do note that the relevant driver here is not yet using DM_USB.
>>>>
>>>> Yes, the driver should be fixed some day. We would really like to take
>>>> this task, but unfortunately, we cannot, at least not right now.
>>>> But we do need that USB working on our board and not only on TI EVMs...
>>>
>>> See above, I believe 042fdb7cabb8270eb86c45f11263fa91c12e3b65 is bogus.
>>
>> Well, agreed.
>>
>>>
>>>> Tom,
>>>> Should we fall back to v1 and have a worse solution for the base addresses?
>>>
>>> OK, I see me being USB custodian has exactly zero value now ... good to
>>> know.
>>
>> I believe we should start fixing things and probably the best time
>> would be... now? I mean, we can't fix things in the past, right?
>>
>
>

-- 
Thanks and regards,
Uri

[U-Boot] [PATCH v3 2/7] arm: usb: dra7xx: xHCI registers based on USB port index

[Marek Vasut]

On 02/22/2017 01:58 PM, Uri Mashiach wrote:

[...]

>> Someone would have to put that effort in.
>>
>> I wonder if instead of defining INDEX0, INDEX1 ... this could be turned
>> into integer ...
>>
> 
> Something like the following?
> 
> #ifdef CONFIG_DRA7XX
> #if CONFIG_USB_XHCI_DRA7XX_INDEX == 1
> ...
> #elif CONFIG_USB_XHCI_DRA7XX_INDEX == 0
> ...
> #endif /* CONFIG_USB_XHCI_DRA7XX_INDEX == 1 */
> #elif defined CONFIG_AM43XX
> ...

Yes


-- 
Best regards,
Marek Vasut