Hello Swami, Yes, davinci_source_power is contolling VBUS power through I2C/pcf8754 via gpio_set_value_cansleep. We do not need to be a usb host, thus disabling this call solved my problem.
Thanks again. David. SS> David, SS> You would have to provide an appropriate way of SS> controlling the charge pump on your board. You would have to do SS> the same in davinci.c file (at this point in time). SS> It seems that we are invoking TI EVM way of controlling SS> the charge pump but that is not the case with you and hence the crash. SS> Regards SS> swami >> -----Original Message----- >> From: [email protected] >> [mailto:[email protected]] On Behalf >> Of David Goshadze >> Sent: Wednesday, March 03, 2010 10:11 PM >> To: [email protected] >> Subject: Re: 2.6.32 musb_hdrc fails to load (DM6446) >> >> Swami, >> >> as I told you calling usb_setup succeeds, but kernel crashes while >> loading musb_hdrc: >> >> >> ************************************************************************** >> ***************** >> >> musb_hdrc: version 6.0, pio, peripheral, debug=0 >> bus: 'platform': really_probe: probing driver musb_hdrc with device >> musb_hdrc >> bus: 'platform': really_probe: probing driver nop_usb_xceiv with device >> nop_usb_xceiv >> BY DTG: __gpio_cansleep called with gpio = 160 >> Unable to handle kernel NULL pointer dereference at virtual address >> 00000020 >> pgd = c1114000 >> [00000020] *pgd=811ba031, *pte=00000000, *ppte=00000000 >> Internal error: Oops: 17 [#1] PREEMPT >> last sysfs file: >> /sys/devices/platform/davinci_nand.0/mtd/mtd4/mtdblock4/removable >> Modules linked in: musb_hdrc(+) nop_usb_xceiv >> CPU: 0 Not tainted (2.6.32 #40) >> PC is at gpio_set_value_cansleep+0x2c/0x5c >> LR is at gpio_set_value_cansleep+0x18/0x5c >> pc : [<c015c27c>] lr : [<c015c268>] psr: 40000013 >> sp : c2b47dc0 ip : 00003411 fp : c02d0b28 >> r10: 00000000 r9 : c02d0b20 r8 : c3c26360 >> r7 : 00147900 r6 : 00000001 r5 : 000000a0 r4 : 00000000 >> r3 : c0304940 r2 : 00000780 r1 : c0294777 r0 : c02947a9 >> Flags: nZcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user >> Control: 0005317f Table: 81114000 DAC: 00000015 >> Process modprobe (pid: 278, stack limit = 0xc2b46270) >> Stack: (0xc2b47dc0 to 0xc2b48000) >> 7dc0: 00000000 bf00b310 fec64000 bf006e20 c1472000 c1472000 fec64000 >> bf00fc18 >> 7de0: c02d0bd8 c1472000 c02d0b00 00000000 c3c26360 bf00f354 c3803180 >> 00000001 >> 7e00: 00000000 00000000 0000000c fec64000 c1472044 c014bcb8 c2b47e6c >> 00000ba0 >> 7e20: a0000013 c3806850 00000000 000000d0 c039dadc c00d1d00 c384b240 >> c3805000 >> 7e40: c3437520 c3437520 c2b47e90 c009db4c c2b47e58 c00d25dc 00000006 >> 17d828dc >> 7e60: c384b240 c00d21c8 c3c27540 c3c27540 c2b47e90 00000000 c3c27540 >> c2b47e90 >> 7e80: c384b240 00000001 00000000 c00d3128 c384b240 c3437520 00000000 >> c02d0b28 >> 7ea0: c02d0b28 bf00b29c c02de300 c3c26360 c2b46000 00000000 00000000 >> c01737b4 >> 7ec0: c02d0b28 c01728dc c3849540 c02d0b28 c2b47ed8 c02d0b28 c02d0b5c >> bf00b29c >> 7ee0: c02de300 c0172a48 00000000 c01729e8 bf00b29c c0172088 c3803638 >> c384a1b0 >> 7f00: 00000001 bf00b29c bf00b29c c01719b0 bf009cb6 bf009cb6 c2b47f18 >> 00000001 >> 7f20: bf00b288 bf00b29c 00000000 c0021064 000160d0 c0172d1c 00000001 >> bf00b288 >> 7f40: bf00f000 00000000 c0021064 c0173a24 00000000 bf00b328 bf00f000 >> c00203ac >> 7f60: 0000c074 bf00b328 00000000 40164000 c0021064 00000000 bf00b328 >> 00000000 >> 7f80: 40164000 c005e294 40164000 0000c074 00016230 0000979c 00000000 >> 00000000 >> 7fa0: 00000080 c0020ee0 0000979c 00000000 40164000 0000c074 00016230 >> 00016230 >> 7fc0: 0000979c 00000000 00000000 00000080 00000000 000160dc 000160d0 >> 00000000 >> 7fe0: 00016240 be916a04 0000b208 400fd7d4 60000010 40164000 e59de040 >> e28dd048 >> [<c015c27c>] (gpio_set_value_cansleep+0x2c/0x5c) from [<bf006e20>] >> (davinci_source_power+0x34/0x4c [musb_hdrc]) >> [<bf006e20>] (davinci_source_power+0x34/0x4c [musb_hdrc]) from >> [<bf00fc18>] (musb_platform_init+0x5c/0xd4 [musb_hdrc]) >> [<bf00fc18>] (musb_platform_init+0x5c/0xd4 [musb_hdrc]) from [<bf00f354>] >> (musb_probe+0x1c8/0xa30 [musb_hdrc]) >> [<bf00f354>] (musb_probe+0x1c8/0xa30 [musb_hdrc]) from [<c01737b4>] >> (platform_drv_probe+0x18/0x1c) >> [<c01737b4>] (platform_drv_probe+0x18/0x1c) from [<c01728dc>] >> (driver_probe_device+0x114/0x220) >> [<c01728dc>] (driver_probe_device+0x114/0x220) from [<c0172a48>] >> (__driver_attach+0x60/0x84) >> [<c0172a48>] (__driver_attach+0x60/0x84) from [<c0172088>] >> (bus_for_each_dev+0x44/0x74) >> [<c0172088>] (bus_for_each_dev+0x44/0x74) from [<c01719b0>] >> (bus_add_driver+0xa8/0x234) >> [<c01719b0>] (bus_add_driver+0xa8/0x234) from [<c0172d1c>] >> (driver_register+0xa8/0x134) >> [<c0172d1c>] (driver_register+0xa8/0x134) from [<c0173a24>] >> (platform_driver_probe+0x18/0x98) >> [<c0173a24>] (platform_driver_probe+0x18/0x98) from [<c00203ac>] >> (do_one_initcall+0x5c/0x1bc) >> [<c00203ac>] (do_one_initcall+0x5c/0x1bc) from [<c005e294>] >> (sys_init_module+0xbc/0x1e8) >> [<c005e294>] (sys_init_module+0xbc/0x1e8) from [<c0020ee0>] >> (ret_fast_syscall+0x0/0x28) >> Code: e0020593 e59f302c e59f002c e7934002 (e5941020) >> bus: 'platform': add driver watchdog >> ---[ end trace 49c8ae5707d7b633 ]--- >> udevd-event[271]: '/sbin/modprobe -b platform:musb_hdrc' abnormal exit >> >> ************************************************************************** >> ***************** >> >> >> As I can see davinci_source_power from musb/davinci.c calls >> gpio_set_value_cansleep while chip structure is not initialized. >> I'll try to disable this call. >> >> >> Thanks. >> David. >> >> >> SS> David, >> SS> I was assuming that you are using TI EVM platform. If >> SS> you are using your own platform you can call the usb_setup >> SS> directly from your board init file. >> >> SS> Regards >> SS> swami >> >> >> -----Original Message----- >> >> From: [email protected] >> >> [mailto:[email protected]] On >> Behalf >> >> Of David Goshadze >> >> Sent: Tuesday, March 02, 2010 5:46 PM >> >> To: [email protected] >> >> Subject: Re: 2.6.32 musb_hdrc fails to load (DM6446) >> >> >> >> Swami, Thanks again for help. >> >> >> >> I compiled I2C Support, though my board doesn't have one. >> >> As I noticed, due to dev->bus->probe failure for I2C devices, >> >> evm_u35_setup is never called. >> >> >> >> Can you suggest some workaround? (calling usb_setup registered USB >> >> devise, but crashed kernel later as expected, due to chip was not >> >> registered). >> >> >> >> David. >> >> >> >> >> >> >> >> SS> David, >> >> >> >> SS> Did you compile in the I2c Expander support in the kernel. I >> >> SS> think you are missing this part of the config. Can you confirm. >> >> >> >> SS> regards >> >> SS> swami >> >> >> >> SS> ________________________________________ >> >> SS> From: [email protected] >> >> SS> [[email protected]] On Behalf >> >> SS> Of David Goshadze [[email protected]] >> >> SS> Sent: Saturday, February 27, 2010 6:05 PM >> >> SS> To: [email protected] >> >> SS> Subject: Re: 2.6.32 musb_hdrc fails to load (DM6446) >> >> >> >> SS> Hi all again. >> >> >> >> SS> Digging into sources I discovered that setup_usb (from >> >> SS> arch/arm/mach-davinci/usb.c) is never called, thus platform device >> >> SS> register for usb_dev is also never called. >> >> >> >> SS> Wondering for a reason. >> >> SS> Please help. >> >> >> >> SS> David. >> >> >> >> SS>> David, >> >> SS>> Could you try the aarago git davinci tree? >> >> >> >> SS>> From the below log it seems there could be a >> >> SS>> configuration issue between the base port and the driver init. >> >> >> >> SS>> Regards >> >> SS>> swami >> >> >> >> >>> -----Original Message----- >> >> >>> From: davinci-linux-open-source- >> >> >>> [email protected] [mailto:davinci- >> linux- >> >> open- >> >> >>> [email protected]] On Behalf Of >> >> David >> >> >>> Goshadze >> >> >>> Sent: Wednesday, February 24, 2010 11:59 PM >> >> >>> To: [email protected] >> >> >>> Subject: 2.6.32 musb_hdrc fails to load (DM6446) >> >> >>> >> >> >>> Hi all! >> >> >>> >> >> >>> Please help with the following problem: >> >> >>> >> >> >>> I successfully loaded an built 2.6.32 for our DM6446 based board. >> >> >>> >> >> >>> Everything's just fine except I cannot load musb_hdrc driver. >> >> >>> I got: >> >> >>> modprobe musb_hdrc : version 6.0, cppi-dma, peripherial, debug=0 >> >> >>> FATAL: Error inserting musb_hdrc >> >> >>> (/lib/modules/2/6/32/kernel/drivers/usb/musb/musb_hdrc.ko): No such >> >> device >> >> >>> nop_usb_xceiv loaded successfully. >> >> >>> >> >> >>> use_dma=n doesn't help. >> >> >>> >> >> >>> Spending 2 days trying various kernel configurations and googling >> >> didn't >> >> >>> help. >> >> >>> >> >> >>> With 2.6.10 (mvl) everything was fine. >> >> >>> >> >> >>> I'm nearly gone mad with this. >> >> >>> >> >> >>> David. >> >> >>> >> >> >>> >> >> >>> >> >> >>> _______________________________________________ >> >> >>> Davinci-linux-open-source mailing list >> >> >>> [email protected] >> >> >>> http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open- >> source >> >> >> >> >> >> >> >> SS> ___________________________ >> >> SS> David Goshadze, >> >> SS> Chief project developer, >> >> SS> PLC Digimeq, >> >> SS> 5, Vozdvizhenka str. >> >> SS> Moscow, Russian Federation >> >> >> >> SS> Ph +7 495 9505231 >> >> SS> Mob +7 916 604 56 93 >> >> SS> Fax +7 495 9505231 >> >> SS> [email protected] >> >> SS> www.digimeq.com >> >> >> >> SS> _______________________________________________ >> >> SS> Davinci-linux-open-source mailing list >> >> SS> [email protected] >> >> SS> http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open- >> source >> >> >> >> >> >> ___________________________ >> >> David Goshadze, >> >> Chief project developer, >> >> PLC Digimeq, >> >> 5, Vozdvizhenka str. >> >> Moscow, Russian Federation >> >> >> >> Ph +7 495 9505231 >> >> Mob +7 916 604 56 93 >> >> Fax +7 495 9505231 >> >> [email protected] >> >> www.digimeq.com >> >> >> >> _______________________________________________ >> >> Davinci-linux-open-source mailing list >> >> [email protected] >> >> http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source >> >> >> >> ___________________________ >> David Goshadze, >> Chief project developer, >> PLC Digimeq, >> 5, Vozdvizhenka str. >> Moscow, Russian Federation >> >> Ph +7 495 9505231 >> Mob +7 916 604 56 93 >> Fax +7 495 9505231 >> [email protected] >> www.digimeq.com >> >> _______________________________________________ >> Davinci-linux-open-source mailing list >> [email protected] >> http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source ___________________________ David Goshadze, Chief project developer, PLC Digimeq, 5, Vozdvizhenka str. Moscow, Russian Federation Ph +7 495 9505231 Mob +7 916 604 56 93 Fax +7 495 9505231 [email protected] www.digimeq.com _______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
