USB broken on Banana Pi in Linux 4.6

USB broken on Banana Pi in Linux 4.6

[Marc Haber]

Hi,

on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
and in 4.6 there is not even /dev/bus/usb.

Here is the log excerpt from a 4.5 kernel coming up:

May 15 09:30:14 cadencia kernel: [    5.307730] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
May 15 09:30:14 cadencia kernel: [    5.312891] ehci-platform: EHCI generic platform driver
May 15 09:30:14 cadencia kernel: [    5.315579] sun4i-ss 1c15000.crypto-engine: no reset control found
May 15 09:30:14 cadencia kernel: [    5.317303] sun4i-ss 1c15000.crypto-engine: Die ID 0
May 15 09:30:14 cadencia kernel: [    5.322742] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver
May 15 09:30:14 cadencia kernel: [    5.332052] ohci-platform: OHCI generic platform driver
May 15 09:30:14 cadencia kernel: [    5.360131] axp20x 0-0034: AXP20x variant AXP209 found
May 15 09:30:14 cadencia kernel: [    5.405989] axp20x 0-0034: AXP20X driver loaded
May 15 09:30:14 cadencia kernel: [    5.409201] ehci-platform 1c14000.usb: EHCI Host Controller
May 15 09:30:14 cadencia kernel: [    5.409271] ehci-platform 1c14000.usb: new USB bus registered, assigned bus number 1
May 15 09:30:14 cadencia kernel: [    5.409506] ehci-platform 1c14000.usb: irq 29, io mem 0x01c14000
May 15 09:30:14 cadencia kernel: [    5.410553] sunxi-wdt 1c20c90.watchdog: Watchdog enabled (timeout=16 sec, nowayout=0)
May 15 09:30:14 cadencia kernel: [    5.420414] ehci-platform 1c14000.usb: USB 2.0 started, EHCI 1.00
May 15 09:30:14 cadencia kernel: [    5.420977] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
May 15 09:30:14 cadencia kernel: [    5.420998] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
May 15 09:30:14 cadencia kernel: [    5.421010] usb usb1: Product: EHCI Host Controller
May 15 09:30:14 cadencia kernel: [    5.421021] usb usb1: Manufacturer: Linux 4.5.4-zgbpi-armmp-lpae ehci_hcd
May 15 09:30:14 cadencia kernel: [    5.421033] usb usb1: SerialNumber: 1c14000.usb
May 15 09:30:14 cadencia kernel: [    5.422317] hub 1-0:1.0: USB hub found
May 15 09:30:14 cadencia kernel: [    5.422431] hub 1-0:1.0: 1 port detected
May 15 09:30:14 cadencia kernel: [    5.423753] ehci-platform 1c1c000.usb: EHCI Host Controller
May 15 09:30:14 cadencia kernel: [    5.423814] ehci-platform 1c1c000.usb: new USB bus registered, assigned bus number 2
May 15 09:30:14 cadencia kernel: [    5.424055] ehci-platform 1c1c000.usb: irq 33, io mem 0x01c1c000
May 15 09:30:14 cadencia kernel: [    5.432424] ehci-platform 1c1c000.usb: USB 2.0 started, EHCI 1.00
May 15 09:30:14 cadencia kernel: [    5.433089] usb usb2: New USB device found, idVendor=1d6b, idProduct=0002
May 15 09:30:14 cadencia kernel: [    5.433110] usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
May 15 09:30:14 cadencia kernel: [    5.433122] usb usb2: Product: EHCI Host Controller
May 15 09:30:14 cadencia kernel: [    5.433133] usb usb2: Manufacturer: Linux 4.5.4-zgbpi-armmp-lpae ehci_hcd
May 15 09:30:14 cadencia kernel: [    5.433144] usb usb2: SerialNumber: 1c1c000.usb
May 15 09:30:14 cadencia kernel: [    5.434472] hub 2-0:1.0: USB hub found
May 15 09:30:14 cadencia kernel: [    5.434595] hub 2-0:1.0: 1 port detected
May 15 09:30:14 cadencia kernel: [    5.436189] ohci-platform 1c14400.usb: Generic Platform OHCI controller
May 15 09:30:14 cadencia kernel: [    5.436528] ohci-platform 1c14400.usb: new USB bus registered, assigned bus number 3
May 15 09:30:14 cadencia kernel: [    5.436779] ohci-platform 1c14400.usb: irq 30, io mem 0x01c14400
May 15 09:30:14 cadencia kernel: [    5.497002] usb usb3: New USB device found, idVendor=1d6b, idProduct=0001
May 15 09:30:14 cadencia kernel: [    5.497032] usb usb3: New USB device strings: Mfr=3, Product=2, SerialNumber=1
May 15 09:30:14 cadencia kernel: [    5.497045] usb usb3: Product: Generic Platform OHCI controller
May 15 09:30:14 cadencia kernel: [    5.497056] usb usb3: Manufacturer: Linux 4.5.4-zgbpi-armmp-lpae ohci_hcd
May 15 09:30:14 cadencia kernel: [    5.497068] usb usb3: SerialNumber: 1c14400.usb

In kernel 4.6, the message "ohci-platform: OHCI generic platform
driver" is the last one, and "ehci-platform 1c14000.usb: EHCI Host
Controller" is the first one that is missing.

Is this already a known issue? Or, does a 4.6 kernel need to be
configured differently if you want USB?

Greetings
Marc



-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany    |  lose things."    Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

Re: USB broken on Banana Pi in Linux 4.6

[Greg KH]

On Mon, May 30, 2016 at 09:02:54PM +0200, Marc Haber wrote:
> Hi,
> 
> on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
> is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
> and in 4.6 there is not even /dev/bus/usb.

Hint, include linux-usb@vger in usb-specific questions...

> 
> Here is the log excerpt from a 4.5 kernel coming up:
> 
> May 15 09:30:14 cadencia kernel: [    5.307730] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
> May 15 09:30:14 cadencia kernel: [    5.312891] ehci-platform: EHCI generic platform driver
> May 15 09:30:14 cadencia kernel: [    5.315579] sun4i-ss 1c15000.crypto-engine: no reset control found
> May 15 09:30:14 cadencia kernel: [    5.317303] sun4i-ss 1c15000.crypto-engine: Die ID 0
> May 15 09:30:14 cadencia kernel: [    5.322742] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver
> May 15 09:30:14 cadencia kernel: [    5.332052] ohci-platform: OHCI generic platform driver
> May 15 09:30:14 cadencia kernel: [    5.360131] axp20x 0-0034: AXP20x variant AXP209 found
> May 15 09:30:14 cadencia kernel: [    5.405989] axp20x 0-0034: AXP20X driver loaded
> May 15 09:30:14 cadencia kernel: [    5.409201] ehci-platform 1c14000.usb: EHCI Host Controller
> May 15 09:30:14 cadencia kernel: [    5.409271] ehci-platform 1c14000.usb: new USB bus registered, assigned bus number 1
> May 15 09:30:14 cadencia kernel: [    5.409506] ehci-platform 1c14000.usb: irq 29, io mem 0x01c14000
> May 15 09:30:14 cadencia kernel: [    5.410553] sunxi-wdt 1c20c90.watchdog: Watchdog enabled (timeout=16 sec, nowayout=0)
> May 15 09:30:14 cadencia kernel: [    5.420414] ehci-platform 1c14000.usb: USB 2.0 started, EHCI 1.00
> May 15 09:30:14 cadencia kernel: [    5.420977] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
> May 15 09:30:14 cadencia kernel: [    5.420998] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> May 15 09:30:14 cadencia kernel: [    5.421010] usb usb1: Product: EHCI Host Controller
> May 15 09:30:14 cadencia kernel: [    5.421021] usb usb1: Manufacturer: Linux 4.5.4-zgbpi-armmp-lpae ehci_hcd
> May 15 09:30:14 cadencia kernel: [    5.421033] usb usb1: SerialNumber: 1c14000.usb
> May 15 09:30:14 cadencia kernel: [    5.422317] hub 1-0:1.0: USB hub found
> May 15 09:30:14 cadencia kernel: [    5.422431] hub 1-0:1.0: 1 port detected
> May 15 09:30:14 cadencia kernel: [    5.423753] ehci-platform 1c1c000.usb: EHCI Host Controller
> May 15 09:30:14 cadencia kernel: [    5.423814] ehci-platform 1c1c000.usb: new USB bus registered, assigned bus number 2
> May 15 09:30:14 cadencia kernel: [    5.424055] ehci-platform 1c1c000.usb: irq 33, io mem 0x01c1c000
> May 15 09:30:14 cadencia kernel: [    5.432424] ehci-platform 1c1c000.usb: USB 2.0 started, EHCI 1.00
> May 15 09:30:14 cadencia kernel: [    5.433089] usb usb2: New USB device found, idVendor=1d6b, idProduct=0002
> May 15 09:30:14 cadencia kernel: [    5.433110] usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> May 15 09:30:14 cadencia kernel: [    5.433122] usb usb2: Product: EHCI Host Controller
> May 15 09:30:14 cadencia kernel: [    5.433133] usb usb2: Manufacturer: Linux 4.5.4-zgbpi-armmp-lpae ehci_hcd
> May 15 09:30:14 cadencia kernel: [    5.433144] usb usb2: SerialNumber: 1c1c000.usb
> May 15 09:30:14 cadencia kernel: [    5.434472] hub 2-0:1.0: USB hub found
> May 15 09:30:14 cadencia kernel: [    5.434595] hub 2-0:1.0: 1 port detected
> May 15 09:30:14 cadencia kernel: [    5.436189] ohci-platform 1c14400.usb: Generic Platform OHCI controller
> May 15 09:30:14 cadencia kernel: [    5.436528] ohci-platform 1c14400.usb: new USB bus registered, assigned bus number 3
> May 15 09:30:14 cadencia kernel: [    5.436779] ohci-platform 1c14400.usb: irq 30, io mem 0x01c14400
> May 15 09:30:14 cadencia kernel: [    5.497002] usb usb3: New USB device found, idVendor=1d6b, idProduct=0001
> May 15 09:30:14 cadencia kernel: [    5.497032] usb usb3: New USB device strings: Mfr=3, Product=2, SerialNumber=1
> May 15 09:30:14 cadencia kernel: [    5.497045] usb usb3: Product: Generic Platform OHCI controller
> May 15 09:30:14 cadencia kernel: [    5.497056] usb usb3: Manufacturer: Linux 4.5.4-zgbpi-armmp-lpae ohci_hcd
> May 15 09:30:14 cadencia kernel: [    5.497068] usb usb3: SerialNumber: 1c14400.usb
> 
> In kernel 4.6, the message "ohci-platform: OHCI generic platform
> driver" is the last one, and "ehci-platform 1c14000.usb: EHCI Host
> Controller" is the first one that is missing.
> 
> Is this already a known issue? Or, does a 4.6 kernel need to be
> configured differently if you want USB?

Are you sure you configured in the correct host controller that you
want?  Have you done a diff of your .config files to see what you
changed?  How did you create your 4.6 config?

thanks,

greg k-h

Re: USB broken on Banana Pi in Linux 4.6

[Marc Haber]

On Mon, May 30, 2016 at 01:47:12PM -0700, Greg KH wrote:
> On Mon, May 30, 2016 at 09:02:54PM +0200, Marc Haber wrote:
> > Hi,
> > 
> > on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
> > is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
> > and in 4.6 there is not even /dev/bus/usb.
> > 
> > In kernel 4.6, the message "ohci-platform: OHCI generic platform
> > driver" is the last one, and "ehci-platform 1c14000.usb: EHCI Host
> > Controller" is the first one that is missing.
> > 
> > Is this already a known issue? Or, does a 4.6 kernel need to be
> > configured differently if you want USB?
> 
> Are you sure you configured in the correct host controller that you
> want?  Have you done a diff of your .config files to see what you
> changed?  How did you create your 4.6 config?

I used make oldconfig, and I diffed the configs for (hci|usb), and
there is no difference.

Greetings
Marc

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany    |  lose things."    Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

Re: USB broken on Banana Pi in Linux 4.6

[Greg KH]

On Fri, Jun 03, 2016 at 08:53:58AM +0200, Marc Haber wrote:
> On Mon, May 30, 2016 at 01:47:12PM -0700, Greg KH wrote:
> > On Mon, May 30, 2016 at 09:02:54PM +0200, Marc Haber wrote:
> > > Hi,
> > > 
> > > on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
> > > is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
> > > and in 4.6 there is not even /dev/bus/usb.
> > > 
> > > In kernel 4.6, the message "ohci-platform: OHCI generic platform
> > > driver" is the last one, and "ehci-platform 1c14000.usb: EHCI Host
> > > Controller" is the first one that is missing.
> > > 
> > > Is this already a known issue? Or, does a 4.6 kernel need to be
> > > configured differently if you want USB?
> > 
> > Are you sure you configured in the correct host controller that you
> > want?  Have you done a diff of your .config files to see what you
> > changed?  How did you create your 4.6 config?
> 
> I used make oldconfig, and I diffed the configs for (hci|usb), and
> there is no difference.

There might be other things than HCI|USB that are the issue here...

Re: USB broken on Banana Pi in Linux 4.6

[Marc Haber]

[-- Attachment #1: Type: text/plain, Size: 1574 bytes --]

On Fri, Jun 03, 2016 at 08:35:11AM -0700, Greg KH wrote:
> On Fri, Jun 03, 2016 at 08:53:58AM +0200, Marc Haber wrote:
> > On Mon, May 30, 2016 at 01:47:12PM -0700, Greg KH wrote:
> > > On Mon, May 30, 2016 at 09:02:54PM +0200, Marc Haber wrote:
> > > > Hi,
> > > > 
> > > > on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
> > > > is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
> > > > and in 4.6 there is not even /dev/bus/usb.
> > > > 
> > > > In kernel 4.6, the message "ohci-platform: OHCI generic platform
> > > > driver" is the last one, and "ehci-platform 1c14000.usb: EHCI Host
> > > > Controller" is the first one that is missing.
> > > > 
> > > > Is this already a known issue? Or, does a 4.6 kernel need to be
> > > > configured differently if you want USB?
> > > 
> > > Are you sure you configured in the correct host controller that you
> > > want?  Have you done a diff of your .config files to see what you
> > > changed?  How did you create your 4.6 config?
> > 
> > I used make oldconfig, and I diffed the configs for (hci|usb), and
> > there is no difference.
> 
> There might be other things than HCI|USB that are the issue here...

Full config diff attached. Hints are appreciated.

Greetings
Marc

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany    |  lose things."    Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

[-- Attachment #2: configdiff --]
[-- Type: text/plain, Size: 5481 bytes --]

--- /boot/config-4.5.4-zgbpi-armmp-lpae	2016-05-13 17:02:25.000000000 +0200
+++ /boot/config-4.6.0-zgbpi-armmp-lpae	2016-05-16 14:18:06.000000000 +0200
@@ -3 +3 @@
-# Linux/arm 4.5.4 Kernel Configuration
+# Linux/arm 4.6.0 Kernel Configuration
@@ -173,0 +174,2 @@
+# CONFIG_KALLSYMS_ABSOLUTE_PERCPU is not set
+CONFIG_KALLSYMS_BASE_RELATIVE=y
@@ -183 +185 @@
-# CONFIG_BPF_SYSCALL is not set
+CONFIG_BPF_SYSCALL=y
@@ -368,0 +371 @@
+# CONFIG_ARCH_ARTPEC is not set
@@ -463 +466,2 @@
-# CONFIG_ARM_KERNMEM_PERMS is not set
+CONFIG_DEBUG_RODATA=y
+CONFIG_DEBUG_ALIGN_RODATA=y
@@ -712 +715,0 @@
-CONFIG_INET_LRO=m
@@ -1126 +1128,0 @@
-CONFIG_NETLINK_MMAP=y
@@ -1167,0 +1170 @@
+# CONFIG_BT_LEDS is not set
@@ -1181,0 +1185 @@
+CONFIG_AF_KCM=m
@@ -1207,0 +1212,3 @@
+CONFIG_DST_CACHE=y
+CONFIG_NET_DEVLINK=m
+CONFIG_MAY_USE_DEVLINK=m
@@ -1234 +1241 @@
-CONFIG_REGMAP_SPI=m
+CONFIG_REGMAP_SPI=y
@@ -1236 +1242,0 @@
-CONFIG_REGMAP_IRQ=y
@@ -1246 +1252 @@
-# CONFIG_ARM_CCI500_PMU is not set
+# CONFIG_ARM_CCI5xx_PMU is not set
@@ -1407,0 +1414 @@
+# CONFIG_PANEL is not set
@@ -1440,0 +1448,4 @@
+# VOP Bus Driver
+#
+
+#
@@ -1454,0 +1466,4 @@
+
+#
+# VOP Driver
+#
@@ -1575,0 +1591 @@
+# CONFIG_MACSEC is not set
@@ -1816,0 +1833 @@
+# CONFIG_TOUCHSCREEN_MELFAS_MIP4 is not set
@@ -1829 +1845,0 @@
-# CONFIG_TOUCHSCREEN_TS4800 is not set
@@ -1856 +1871,0 @@
-# CONFIG_INPUT_AXP20X_PEK is not set
@@ -1869,0 +1885 @@
+# CONFIG_RMI4_CORE is not set
@@ -1922 +1937,0 @@
-# CONFIG_SERIAL_8250_INGENIC is not set
@@ -1951,0 +1967 @@
+# CONFIG_SERIAL_MVEBU_UART is not set
@@ -1981,0 +1998 @@
+# CONFIG_I2C_DEMUX_PINCTRL is not set
@@ -2031,0 +2049 @@
+# CONFIG_SPI_AXI_SPI_ENGINE is not set
@@ -2034,0 +2053 @@
+# CONFIG_SPI_DESIGNWARE is not set
@@ -2048 +2066,0 @@
-# CONFIG_SPI_DESIGNWARE is not set
@@ -2097 +2115 @@
-CONFIG_PINCTRL_SUNXI_COMMON=y
+CONFIG_PINCTRL_SUNXI=y
@@ -2109,0 +2128 @@
+CONFIG_PINCTRL_SUN8I_H3_R=y
@@ -2131,0 +2151 @@
+# CONFIG_GPIO_MPC8XXX is not set
@@ -2147,0 +2168 @@
+# CONFIG_GPIO_TPIC2810 is not set
@@ -2159,0 +2181 @@
+# CONFIG_GPIO_PISOSR is not set
@@ -2195 +2216,0 @@
-CONFIG_AXP20X_POWER=y
@@ -2246,0 +2268 @@
+# CONFIG_SENSORS_LTC2990 is not set
@@ -2364 +2385,0 @@
-# CONFIG_TS4800_WATCHDOG is not set
@@ -2366 +2386,0 @@
-# CONFIG_BCM7038_WDT is not set
@@ -2389,0 +2410 @@
+# CONFIG_MFD_ACT8945A is not set
@@ -2397 +2418,2 @@
-CONFIG_MFD_AXP20X=y
+# CONFIG_MFD_AXP20X_I2C is not set
+# CONFIG_MFD_AXP20X_RSB is not set
@@ -2454,0 +2477 @@
+# CONFIG_MFD_TPS65086 is not set
@@ -2460 +2482,0 @@
-# CONFIG_MFD_TPS65912 is not set
@@ -2491 +2512,0 @@
-# CONFIG_REGULATOR_AXP20X is not set
@@ -2566,0 +2588,5 @@
+# ACP (Audio CoProcessor) Configuration
+#
+# CONFIG_DRM_AMD_ACP is not set
+
+#
@@ -2636,0 +2663 @@
+CONFIG_SND_JACK_INPUT_DEV=y
@@ -2702,0 +2730 @@
+# CONFIG_SND_SUN4I_SPDIF is not set
@@ -2732 +2760,2 @@
-# CONFIG_SND_SOC_PCM179X is not set
+# CONFIG_SND_SOC_PCM179X_I2C is not set
+# CONFIG_SND_SOC_PCM179X_SPI is not set
@@ -2736,0 +2766 @@
+# CONFIG_SND_SOC_RT5616 is not set
@@ -2801,0 +2832 @@
+# CONFIG_HID_CMEDIA is not set
@@ -3153,0 +3185 @@
+# CONFIG_LEDS_IS31FL32XX is not set
@@ -3208 +3239,0 @@
-# CONFIG_RTC_DRV_DS3232 is not set
@@ -3216 +3246,0 @@
-# CONFIG_RTC_DRV_PCF2127 is not set
@@ -3218 +3247,0 @@
-# CONFIG_RTC_DRV_PCF8563 is not set
@@ -3219,0 +3249 @@
+# CONFIG_RTC_DRV_PCF8563 is not set
@@ -3242 +3271,0 @@
-# CONFIG_RTC_DRV_MAX6902 is not set
@@ -3243,0 +3273,2 @@
+# CONFIG_RTC_DRV_RX4581 is not set
+# CONFIG_RTC_DRV_RX6110 is not set
@@ -3245 +3276 @@
-# CONFIG_RTC_DRV_DS3234 is not set
+# CONFIG_RTC_DRV_MAX6902 is not set
@@ -3247 +3277,0 @@
-# CONFIG_RTC_DRV_RX4581 is not set
@@ -3248,0 +3279,7 @@
+CONFIG_RTC_I2C_AND_SPI=y
+
+#
+# SPI and I2C RTC drivers
+#
+# CONFIG_RTC_DRV_DS3232 is not set
+# CONFIG_RTC_DRV_PCF2127 is not set
@@ -3298,0 +3336,2 @@
+# CONFIG_QCOM_HIDMA_MGMT is not set
+# CONFIG_QCOM_HIDMA is not set
@@ -3324 +3362,0 @@
-# CONFIG_PANEL is not set
@@ -3343 +3380,0 @@
-# CONFIG_TOUCHSCREEN_SYNAPTICS_I2C_RMI4 is not set
@@ -3350 +3386,0 @@
-# CONFIG_WIMAX_GDM72XX is not set
@@ -3353,2 +3389 @@
-# CONFIG_LUSTRE_FS is not set
-# CONFIG_DGAP is not set
+# CONFIG_LNET is not set
@@ -3358,0 +3394 @@
+# CONFIG_GOLDFISH is not set
@@ -3372,0 +3409 @@
+# CONFIG_COMMON_CLK_CDCE706 is not set
@@ -3379 +3415,0 @@
-# CONFIG_COMMON_CLK_CDCE706 is not set
@@ -3417,0 +3454 @@
+# CONFIG_IOMMU_IO_PGTABLE_ARMV7S is not set
@@ -3455 +3491,0 @@
-# CONFIG_EXTCON_AXP288 is not set
@@ -3475 +3510,0 @@
-# CONFIG_TS4800_IRQ is not set
@@ -3504 +3539,2 @@
-# CONFIG_NVMEM is not set
+CONFIG_NVMEM=m
+CONFIG_NVMEM_SUNXI_SID=m
@@ -3506,2 +3541,0 @@
-# CONFIG_STM_DUMMY is not set
-# CONFIG_STM_SOURCE_CONSOLE is not set
@@ -3553,0 +3588 @@
+CONFIG_FS_ENCRYPTION=m
@@ -3590,0 +3626 @@
+# CONFIG_FAT_DEFAULT_UTF8 is not set
@@ -3608,0 +3645 @@
+# CONFIG_ORANGEFS_FS is not set
@@ -3654 +3691,3 @@
-# CONFIG_NFSD_PNFS is not set
+CONFIG_NFSD_PNFS=y
+CONFIG_NFSD_BLOCKLAYOUT=y
+CONFIG_NFSD_SCSILAYOUT=y
@@ -3763,0 +3803,2 @@
+# CONFIG_PAGE_POISONING is not set
+# CONFIG_DEBUG_PAGE_REF is not set
@@ -3829,0 +3871 @@
+# CONFIG_CPU_HOTPLUG_STATE_CONTROL is not set
@@ -3861,0 +3904 @@
+CONFIG_BPF_EVENTS=y
@@ -3887,0 +3931 @@
+# CONFIG_TEST_BITMAP is not set
@@ -3922 +3966 @@
-# CONFIG_ENCRYPTED_KEYS is not set
+CONFIG_ENCRYPTED_KEYS=m
@@ -3959,2 +4002,0 @@
-CONFIG_CRYPTO_PCOMP=m
-CONFIG_CRYPTO_PCOMP2=y
@@ -4055 +4096,0 @@
-CONFIG_CRYPTO_ZLIB=m
@@ -4080 +4120,0 @@
-CONFIG_PUBLIC_KEY_ALGO_RSA=m

Re: USB broken on Banana Pi in Linux 4.6

[Greg KH]

On Tue, Jun 07, 2016 at 07:06:00PM +0200, Marc Haber wrote:
> On Fri, Jun 03, 2016 at 08:35:11AM -0700, Greg KH wrote:
> > On Fri, Jun 03, 2016 at 08:53:58AM +0200, Marc Haber wrote:
> > > On Mon, May 30, 2016 at 01:47:12PM -0700, Greg KH wrote:
> > > > On Mon, May 30, 2016 at 09:02:54PM +0200, Marc Haber wrote:
> > > > > Hi,
> > > > > 
> > > > > on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
> > > > > is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
> > > > > and in 4.6 there is not even /dev/bus/usb.
> > > > > 
> > > > > In kernel 4.6, the message "ohci-platform: OHCI generic platform
> > > > > driver" is the last one, and "ehci-platform 1c14000.usb: EHCI Host
> > > > > Controller" is the first one that is missing.
> > > > > 
> > > > > Is this already a known issue? Or, does a 4.6 kernel need to be
> > > > > configured differently if you want USB?
> > > > 
> > > > Are you sure you configured in the correct host controller that you
> > > > want?  Have you done a diff of your .config files to see what you
> > > > changed?  How did you create your 4.6 config?
> > > 
> > > I used make oldconfig, and I diffed the configs for (hci|usb), and
> > > there is no difference.
> > 
> > There might be other things than HCI|USB that are the issue here...
> 
> Full config diff attached. Hints are appreciated.

Nothing obvious, can you use 'git bisect' to go from 4.5.0 to 4.6.0 to
find the offending commit?

thanks,

greg k-h

Re: USB broken on Banana Pi in Linux 4.6

[Marc Haber]

On Tue, Jun 07, 2016 at 10:30:17AM -0700, Greg KH wrote:
> Nothing obvious, can you use 'git bisect' to go from 4.5.0 to 4.6.0 to
> find the offending commit?

I can. The first round of bisecting let me end up with
c8b710b3e4348119924051551b836c94835331b1 as the first bad commit,
which is wrong, since c8b710b3e4348119924051551b836c94835331b1^ is bad
as well. I am not sure whether things went well since I had to use git
bisect skip twice because the resulting kernel wouldn't boot on the pi.

A second round of bisecting left me in limbo, since I do not
understand this:

[45/544]mh@fan[zgchroot kernel64][debian_chroot sid_kernel64]:~/linux/debug/linux$ git bisect reset
HEAD is now at 2dcd0af... Linux 4.6
[46/545]mh@fan[zgchroot kernel64][debian_chroot sid_kernel64]:~/linux/debug/linux$ git bisect start
[47/546]mh@fan[zgchroot kernel64][debian_chroot sid_kernel64]:~/linux/debug/linux$ git bisect bad
[48/547]mh@fan[zgchroot kernel64][debian_chroot sid_kernel64]:~/linux/debug/linux$ git bisect bad c8b710b3e4348119924051551b836c94835331b1^
[49/548]mh@fan[zgchroot kernel64][debian_chroot sid_kernel64]:~/linux/debug/linux$ git bisect good v4.5
Some good revs are not ancestor of the bad rev.
git bisect cannot work properly in this case.
Maybe you mistook good and bad revs?
[50/549]mh@fan[zgchroot kernel64][debian_chroot sid_kernel64]:~/linux/debug/linux$

Do I need to start over completely?

Greetings
Marc

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany    |  lose things."    Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

Re: USB broken on Banana Pi in Linux 4.6

[Marc Haber]

On Sat, Jun 11, 2016 at 02:55:04PM +0200, Marc Haber wrote:
> On Tue, Jun 07, 2016 at 10:30:17AM -0700, Greg KH wrote:
> > Nothing obvious, can you use 'git bisect' to go from 4.5.0 to 4.6.0 to
> > find the offending commit?
> 
> I can. The first round of bisecting let me end up with
> c8b710b3e4348119924051551b836c94835331b1 as the first bad commit,
> which is wrong, since c8b710b3e4348119924051551b836c94835331b1^ is bad
> as well. I am not sure whether things went well since I had to use git
> bisect skip twice because the resulting kernel wouldn't boot on the pi.

The kernel panic on boot is caused by bugs in the parport part. I
worked around these by disabling PARPORT in the kernel configuration.

However, a weekend of bisecting just sent me back to commit
d85ce830eef6c10d1e9617172dea4681f02b8424, which is a purely cosmetic
commit. What totally confuses me is the sheer size of the diff.

[8/506]mh@fan:~/linux/debug/linux.bad$ less .git/BISECT_LOG [9/507]mh@fan:~/linux/debug/linux.bad$ git log v4.5..d85ce830eef6c10d1e9617172dea4681f02b8424 | cat
d85ce830eef6c10d1e9617172dea4681f02b8424 perf pmu: Fix misleadingly indented assignment (whitespace)
[10/508]mh@fan:~/linux/debug/linux.bad$ git diff v4.5..d85ce830eef6c10d1e9617172dea4681f02b8424 | wc -l
811131
[11/509]mh@fan:~/linux/debug/linux.bad$ git show d85ce830eef6c10d1e9617172dea4681f02b8424 | wc -l
14
[12/510]mh@fan:~/linux/debug/linux.bad$

Why do I get a 800000+ line diff for a 14 line commit?

This can't be correct. Hints?

Here is the BISECT_LOG:

git bisect start
# bad: [2dcd0af568b0cf583645c8a317dd12e344b1c72a] Linux 4.6
git bisect bad 2dcd0af568b0cf583645c8a317dd12e344b1c72a
# good: [b562e44f507e863c6792946e4e1b1449fbbac85d] Linux 4.5
git bisect good b562e44f507e863c6792946e4e1b1449fbbac85d
# bad: [6b5f04b6cf8ebab9a65d9c0026c650bb2538fd0f] Merge branch 'for-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup
git bisect bad 6b5f04b6cf8ebab9a65d9c0026c650bb2538fd0f
# bad: [96b9b1c95660d4bc5510c5d798d3817ae9f0b391] Merge tag 'tty-4.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
git bisect bad 96b9b1c95660d4bc5510c5d798d3817ae9f0b391
# bad: [277edbabf6fece057b14fb6db5e3a34e00f42f42] Merge tag 'pm+acpi-4.6-rc1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
git bisect bad 277edbabf6fece057b14fb6db5e3a34e00f42f42
# bad: [5ca5446ec5ba5e79a6f271cd026bb153d6850fcc] Merge tag 'pinctrl-v4.6-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
git bisect bad 5ca5446ec5ba5e79a6f271cd026bb153d6850fcc
# bad: [e71c2c1eeb8de7a083a728c5b7e0b83ed1faf047] Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
git bisect bad e71c2c1eeb8de7a083a728c5b7e0b83ed1faf047
# bad: [54fbad54ebcde9db9c7459e9e379f2350c25e1f1] perf mem record: Check for memory events support
git bisect bad 54fbad54ebcde9db9c7459e9e379f2350c25e1f1
# bad: [598b7c6919c7bbcc1243009721a01bc12275ff3e] perf jit: add source line info support
git bisect bad 598b7c6919c7bbcc1243009721a01bc12275ff3e
# bad: [3848c23b19e07188bfa15e3d9a2ac27692f2ff3c] perf report: Don't show blank lines if entry has no callchain
git bisect bad 3848c23b19e07188bfa15e3d9a2ac27692f2ff3c
# bad: [5ac76283b32b116c58e362e99542182ddcfc8262] perf cpumap: Auto initialize cpu__max_{node,cpu}
git bisect bad 5ac76283b32b116c58e362e99542182ddcfc8262
# bad: [cfd92dadc5e830268036efb25ff41618f29c3306] perf sort: Provide a way to find out if per-thread bucketing is in place
git bisect bad cfd92dadc5e830268036efb25ff41618f29c3306
# bad: [3379e0c3effa87d7734fc06277a7023292aadb0c] perf tools: Document the perf sysctls
git bisect bad 3379e0c3effa87d7734fc06277a7023292aadb0c
# bad: [86a2cf3123bfec118bfb98728d88be0668779b2b] perf stat: Making several helper functions static
git bisect bad 86a2cf3123bfec118bfb98728d88be0668779b2b
# bad: [403567217d3fa5d4801f820317ada52e5c5f0e53] perf symbols: Do not read symbols/data from device files
git bisect bad 403567217d3fa5d4801f820317ada52e5c5f0e53
# bad: [d85ce830eef6c10d1e9617172dea4681f02b8424] perf pmu: Fix misleadingly indented assignment (whitespace)
git bisect bad d85ce830eef6c10d1e9617172dea4681f02b8424
# first bad commit: [d85ce830eef6c10d1e9617172dea4681f02b8424] perf pmu: Fix misleadingly indented assignment (whitespace)
[13/511]mh@fan:~/linux/debug/linux.bad$

(started with git checkout v4.6, git bisect start, git bisect bad, git
bisect good v4.5).

Greetings
Marc


-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany    |  lose things."    Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

Re: USB broken on Banana Pi in Linux 4.6

[Alan Stern]

On Sun, 12 Jun 2016, Marc Haber wrote:

> On Sat, Jun 11, 2016 at 02:55:04PM +0200, Marc Haber wrote:
> > On Tue, Jun 07, 2016 at 10:30:17AM -0700, Greg KH wrote:
> > > Nothing obvious, can you use 'git bisect' to go from 4.5.0 to 4.6.0 to
> > > find the offending commit?
> > 
> > I can. The first round of bisecting let me end up with
> > c8b710b3e4348119924051551b836c94835331b1 as the first bad commit,
> > which is wrong, since c8b710b3e4348119924051551b836c94835331b1^ is bad
> > as well. I am not sure whether things went well since I had to use git
> > bisect skip twice because the resulting kernel wouldn't boot on the pi.
> 
> The kernel panic on boot is caused by bugs in the parport part. I
> worked around these by disabling PARPORT in the kernel configuration.
> 
> However, a weekend of bisecting just sent me back to commit
> d85ce830eef6c10d1e9617172dea4681f02b8424, which is a purely cosmetic
> commit. What totally confuses me is the sheer size of the diff.
> 
> [8/506]mh@fan:~/linux/debug/linux.bad$ less .git/BISECT_LOG [9/507]mh@fan:~/linux/debug/linux.bad$ git log v4.5..d85ce830eef6c10d1e9617172dea4681f02b8424 | cat
> d85ce830eef6c10d1e9617172dea4681f02b8424 perf pmu: Fix misleadingly indented assignment (whitespace)
> [10/508]mh@fan:~/linux/debug/linux.bad$ git diff v4.5..d85ce830eef6c10d1e9617172dea4681f02b8424 | wc -l
> 811131
> [11/509]mh@fan:~/linux/debug/linux.bad$ git show d85ce830eef6c10d1e9617172dea4681f02b8424 | wc -l
> 14
> [12/510]mh@fan:~/linux/debug/linux.bad$
> 
> Why do I get a 800000+ line diff for a 14 line commit?
> 
> This can't be correct. Hints?

That diff is the difference all the way from v4.5 to the d85ce830eef6
commit.  Not just the commit itself, but also everything along the way.

> Here is the BISECT_LOG:
> 
> git bisect start
> # bad: [2dcd0af568b0cf583645c8a317dd12e344b1c72a] Linux 4.6
> git bisect bad 2dcd0af568b0cf583645c8a317dd12e344b1c72a
> # good: [b562e44f507e863c6792946e4e1b1449fbbac85d] Linux 4.5
> git bisect good b562e44f507e863c6792946e4e1b1449fbbac85d
> # bad: [6b5f04b6cf8ebab9a65d9c0026c650bb2538fd0f] Merge branch 'for-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup
> git bisect bad 6b5f04b6cf8ebab9a65d9c0026c650bb2538fd0f
> # bad: [96b9b1c95660d4bc5510c5d798d3817ae9f0b391] Merge tag 'tty-4.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
> git bisect bad 96b9b1c95660d4bc5510c5d798d3817ae9f0b391
> # bad: [277edbabf6fece057b14fb6db5e3a34e00f42f42] Merge tag 'pm+acpi-4.6-rc1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
> git bisect bad 277edbabf6fece057b14fb6db5e3a34e00f42f42
> # bad: [5ca5446ec5ba5e79a6f271cd026bb153d6850fcc] Merge tag 'pinctrl-v4.6-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
> git bisect bad 5ca5446ec5ba5e79a6f271cd026bb153d6850fcc
> # bad: [e71c2c1eeb8de7a083a728c5b7e0b83ed1faf047] Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
> git bisect bad e71c2c1eeb8de7a083a728c5b7e0b83ed1faf047
> # bad: [54fbad54ebcde9db9c7459e9e379f2350c25e1f1] perf mem record: Check for memory events support
> git bisect bad 54fbad54ebcde9db9c7459e9e379f2350c25e1f1
> # bad: [598b7c6919c7bbcc1243009721a01bc12275ff3e] perf jit: add source line info support
> git bisect bad 598b7c6919c7bbcc1243009721a01bc12275ff3e
> # bad: [3848c23b19e07188bfa15e3d9a2ac27692f2ff3c] perf report: Don't show blank lines if entry has no callchain
> git bisect bad 3848c23b19e07188bfa15e3d9a2ac27692f2ff3c
> # bad: [5ac76283b32b116c58e362e99542182ddcfc8262] perf cpumap: Auto initialize cpu__max_{node,cpu}
> git bisect bad 5ac76283b32b116c58e362e99542182ddcfc8262
> # bad: [cfd92dadc5e830268036efb25ff41618f29c3306] perf sort: Provide a way to find out if per-thread bucketing is in place
> git bisect bad cfd92dadc5e830268036efb25ff41618f29c3306
> # bad: [3379e0c3effa87d7734fc06277a7023292aadb0c] perf tools: Document the perf sysctls
> git bisect bad 3379e0c3effa87d7734fc06277a7023292aadb0c
> # bad: [86a2cf3123bfec118bfb98728d88be0668779b2b] perf stat: Making several helper functions static
> git bisect bad 86a2cf3123bfec118bfb98728d88be0668779b2b
> # bad: [403567217d3fa5d4801f820317ada52e5c5f0e53] perf symbols: Do not read symbols/data from device files
> git bisect bad 403567217d3fa5d4801f820317ada52e5c5f0e53
> # bad: [d85ce830eef6c10d1e9617172dea4681f02b8424] perf pmu: Fix misleadingly indented assignment (whitespace)
> git bisect bad d85ce830eef6c10d1e9617172dea4681f02b8424
> # first bad commit: [d85ce830eef6c10d1e9617172dea4681f02b8424] perf pmu: Fix misleadingly indented assignment (whitespace)
> [13/511]mh@fan:~/linux/debug/linux.bad$
> 
> (started with git checkout v4.6, git bisect start, git bisect bad, git
> bisect good v4.5).

According to this log, all the commits you tested were bad.  The only 
good commit listed is v4.5 itself.  That doesn't seem right.

Alan Stern

Re: USB broken on Banana Pi in Linux 4.6 [solved]

[Marc Haber]

Hi,

On Mon, May 30, 2016 at 09:02:54PM +0200, Marc Haber wrote:
> on my Bananapis, in kernel 4.6 USB does not work. Kernel configuration
> is USB-wise identical to 4.5 (grepped for differences in (hci|usb)),
> and in 4.6 there is not even /dev/bus/usb.

This turned out to be a configuration issue. 4.6 kernels on Banana Pi
need CONFIG_AXP20X_POWER for working USB. If that driver is missing,
one gets a silent fail.

Thanks for all your help.

Greetings
Marc

-- 
-----------------------------------------------------------------------------
Marc Haber         | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany    |  lose things."    Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421

Re: USB broken on Banana Pi in Linux 4.6

[Sudip Mukherjee]

On Sun, Jun 12, 2016 at 06:47:56PM +0200, Marc Haber wrote:
> On Sat, Jun 11, 2016 at 02:55:04PM +0200, Marc Haber wrote:
> > On Tue, Jun 07, 2016 at 10:30:17AM -0700, Greg KH wrote:
> > > Nothing obvious, can you use 'git bisect' to go from 4.5.0 to 4.6.0 to
> > > find the offending commit?
> > 
> > I can. The first round of bisecting let me end up with
> > c8b710b3e4348119924051551b836c94835331b1 as the first bad commit,
> > which is wrong, since c8b710b3e4348119924051551b836c94835331b1^ is bad
> > as well. I am not sure whether things went well since I had to use git
> > bisect skip twice because the resulting kernel wouldn't boot on the pi.
> 
> The kernel panic on boot is caused by bugs in the parport part. I
> worked around these by disabling PARPORT in the kernel configuration.

May I ask which bug in parport are you reffering to? There was a bad
commit in ppdev which caused boot failure and that was reverted.
But in parport??

regards
sudip