Linuxppc-embedded Digest, Vol 50, Issue 12

[Duy-Ky Nguyen]

Dear Scott,

Once again, you're absolute right, It works so great.

You've helped me out a lot, Scott.

I'm unable to find any word to show my most sincere appreciation to your 
invaluable help.

Very Best Regards,

Duy-Ky

----- Original Message ----- 
From: <linuxppc-embedded-request at ozlabs.org>
To: <linuxppc-embedded at ozlabs.org>
Sent: Wednesday, October 08, 2008 6:00 PM
Subject: Linuxppc-embedded Digest, Vol 50, Issue 12


> Send Linuxppc-embedded mailing list submissions to
> linuxppc-embedded at ozlabs.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://ozlabs.org/mailman/listinfo/linuxppc-embedded
> or, via email, send a message with subject or body 'help' to
> linuxppc-embedded-request at ozlabs.org
>
> You can reach the person managing the list at
> linuxppc-embedded-owner at ozlabs.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Linuxppc-embedded digest..."
>
>
> Today's Topics:
>
>   1. Re: Linuxppc-embedded Digest, Vol 50, Issue 8 (Duy-Ky Nguyen)
>   2. Failure in re-load module (Duy-Ky Nguyen)
>   3. Re: Failure in re-load module (Scott Wood)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 8 Oct 2008 12:53:10 -0700
> From: "Duy-Ky Nguyen" <duykynguyen at hotmail.com>
> Subject: Re: Linuxppc-embedded Digest, Vol 50, Issue 8
> To: <linuxppc-embedded at ozlabs.org>
> Message-ID: <BLU119-DAV13C60547C53073676EDCA7BB3B0 at phx.gbl>
> Content-Type: text/plain; format=flowed; charset="iso-8859-1";
> reply-type=original
>
> Dear Seva,
>
> You're very grrrrrrrrrrrrrrrrrreat, You're absolutely rrrrrrrrrrrrrright,
> Selva
>
> It's very logic, but it forgave me for quite long tong time until now.
>
> Million Thanks and Very Best Wihses to you Selva,
>
> Duy-Ky
>
> ----- Original Message ----- 
> From: <linuxppc-embedded-request at ozlabs.org>
> To: <linuxppc-embedded at ozlabs.org>
> Sent: Tuesday, October 07, 2008 6:00 PM
> Subject: Linuxppc-embedded Digest, Vol 50, Issue 8
>
>
>> Send Linuxppc-embedded mailing list submissions to
>> linuxppc-embedded at ozlabs.org
>>
>> To subscribe or unsubscribe via the World Wide Web, visit
>> https://ozlabs.org/mailman/listinfo/linuxppc-embedded
>> or, via email, send a message with subject or body 'help' to
>> linuxppc-embedded-request at ozlabs.org
>>
>> You can reach the person managing the list at
>> linuxppc-embedded-owner at ozlabs.org
>>
>> When replying, please edit your Subject line so it is more specific
>> than "Re: Contents of Linuxppc-embedded digest..."
>>
>>
>> Today's Topics:
>>
>>   1. CPIO image date problem (Fisherss)
>>   2. Trying to vfree() nonexistent vm area (Duy-Ky Nguyen)
>>   3. German keyboard layout (Schmid Alexander)
>>   4. Re: Trying to vfree() nonexistent vm area (selvamuthukumar v)
>>   5. SysACE driver and Ethernet support? - mainline Linux 2.6.10
>>      and Xilinx ML310 (Juliana Su)
>>
>>
>> ----------------------------------------------------------------------
>>
>> Message: 1
>> Date: Tue, 7 Oct 2008 13:15:17 +0800
>> From: Fisherss <fisherssml at gmail.com>
>> Subject: CPIO image date problem
>> To: linuxppc-embedded at ozlabs.org
>> Message-ID:
>> <25fd1f750810062215q7f33664eo99908ab7dd26f502 at mail.gmail.com>
>> Content-Type: text/plain; charset=ISO-8859-1
>>
>> Hi all
>>
>> I have a weird problem on MPC8548E with Linux Kernel 2.6.24:
>>
>> I use a CPIO image as my embedded system's root file system, but after
>> booting, all files and directories' "date" attribute are set to "Jan
>> 1, 1970".
>>
>> Now what I have got is:
>> ====================
>> U-boot and the cpio image are cleared. The roofs(cpio image) is loaded
>> by Linux Kernel, not U-boot, and the image itself has successfully
>> passed test, so we suppose the problem is in Linux Kernel.
>>
>> The Kernel loading sequence is : time_init() -> cpio image loading ->
>> RTC driver loading
>>
>> That means, at the cpio image loading time, we don't have the RTC
>> driver. In fact, Kernel uses the time_base register instead of RTC(at
>> least at this time). All standard cpio images contain the "date" file
>> attribute, but seems like Kernel just ignores it and uses its own
>> "clock source"(and this clock source should be binded to time_base
>> register).
>>
>> I have located the cpio image unpacking code in Kernel, but can not
>> find anything about date attribute. This part of code is
>> platform-independent.
>>
>> Anyone any idea ??
>>
>> Thanks
>>
>> -- 
>> Fisherss
>> fisherss at gmail.com
>>
>>
>> ------------------------------
>>
>> Message: 2
>> Date: Mon, 6 Oct 2008 23:14:01 -0700
>> From: "Duy-Ky Nguyen" <duykynguyen at hotmail.com>
>> Subject: Trying to vfree() nonexistent vm area
>> To: <linuxppc-embedded at ozlabs.org>
>> Cc: Scott Wood <scottwood at freescale.com>
>> Message-ID: <BLU119-DAV245BF96B99DC961FFF9EFBB380 at phx.gbl>
>> Content-Type: text/plain; format=flowed; charset="iso-8859-1";
>> reply-type=original
>>
>> Hello,
>>
>> My driver uses ioremap() the IMMR_BASE @ 0xE0000000 to access registers.
>> A user-space app opens and close the deive file on every access to
>> register.
>>
>> There's no problem with FreeScale Linuc BSP 2..6.20 for eva board
>> MPC83813E-RDB.
>>
>> For the latest FreeScale Linuc BSP 2..6.24 , I have NO problem for the
>> very
>> FIRST run.
>> But after the 1st run, I have problem with the new FreeScale Linuc BSP
>> 2..6.24 with the following error message
>>
>>        Trying to vfree() nonexistent vm area (c98a0000)
>>
>> where c98a0000 is the mapped address of IMMR_BASE 0xE0000000.
>>
>> So, it appears to me that the device file could be closed improperly as
>> it's
>> closed after the 1st run.
>>
>> I really appreciate your time and help
>>
>> Best Regards,
>>
>> Duy-Ky
>>
>> ///////////////////////
>> Below is outline of the driver
>>
>> init:
>>     ppc_base = (unsigned long) ioremap_nocache(PPC_BASE, PPC_SIZE);
>>
>> //     ppc_base = (unsigned long) ioremap(PPC_BASE, PPC_SIZE);    //
>> failure
>> with this way too
>>
>>
>> close:
>>    iounmap((void __iomem *)ppc_base);
>> //    iounmap(PPC_BASE);        // failure with this way too !
>>
>>
>>
>>
>>
>>
>> ------------------------------
>>
>> Message: 3
>> Date: Tue, 7 Oct 2008 08:35:32 +0200
>> From: "Schmid Alexander" <a.schmid at systeme-steuerungen.de>
>> Subject: German keyboard layout
>> To: <linuxppc-embedded at ozlabs.org>
>> Cc: PHYTEC Support <support at phytec.de>
>> Message-ID:
>> <FCELKENEBBOLLHKPHCCBIELFCEAA.a.schmid at systeme-steuerungen.de>
>> Content-Type: text/plain;   charset="iso-8859-1"
>>
>>
>> Hello, I have linux 2.6.23.1 on a MPC5200B with a USB keyboard!
>> But the default language is US. How can I have a german layout?
>> I tried cross-compiling the package kbd-1.0.6, but this seems to be very
>> tricky! I didn`t finish that.
>>
>> Then I tried it with the loadkmap from the busybox! But there is always
>> the
>> error: /dev/vc0: No such file or directory!
>>
>> Can anyone help me please? I would only need a german keyboard layout!
>>
>> Thank you
>>
>>
>>
>>
>>
>> Alexander Schmid
>> Dipl. Ing. (FH)
>> -Entwicklung-
>> Systeme & Steuerungen GmbH
>> Josef-Buchinger-Strasse 8
>> DE-94481 Grafenau
>>
>>
>>
>>
>> ------------------------------
>>
>> Message: 4
>> Date: Tue, 7 Oct 2008 20:21:13 +0530
>> From: "selvamuthukumar v" <v.selvamuthukumar at gmail.com>
>> Subject: Re: Trying to vfree() nonexistent vm area
>> To: "Duy-Ky Nguyen" <duykynguyen at hotmail.com>
>> Cc: Scott Wood <scottwood at freescale.com>, linuxppc-embedded at ozlabs.org
>> Message-ID:
>> <790896390810070751u71d34ad4t82921a9bb2aaa698 at mail.gmail.com>
>> Content-Type: text/plain; charset=ISO-8859-1
>>
>>> ///////////////////////
>>> Below is outline of the driver
>>>
>>> init:
>>>    ppc_base = (unsigned long) ioremap_nocache(PPC_BASE, PPC_SIZE);
>>>
>>> //     ppc_base = (unsigned long) ioremap(PPC_BASE, PPC_SIZE);    //
>>> failure
>>> with this way too
>>>
>>>
>>> close:
>>>   iounmap((void __iomem *)ppc_base);
>>> //    iounmap(PPC_BASE);        // failure with this way too !
>>>
>>>
>> init and close seems to be wrong combinations. You can try,
>>
>> open - ioremap
>> close - iounmap
>>
>> or
>>
>> init - ioremap
>> exit - iounmap
>>
>> -- 
>> Selva
>>
>>
>> ------------------------------
>>
>> Message: 5
>> Date: Tue, 07 Oct 2008 11:52:09 -0400
>> From: Juliana Su <js084 at bucknell.edu>
>> Subject: SysACE driver and Ethernet support? - mainline Linux 2.6.10
>> and Xilinx ML310
>> To: linuxppc-embedded at ozlabs.org
>> Message-ID: <48EB85A9.1000302 at bucknell.edu>
>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>>
>> Hi All,
>>
>> I finally got PCI support on my Xilinx ML310 with a mainline Linux
>> 2.6.10 kernel and gcc-3.4.5-glibc-2.3.6 (with the help of the awesome
>> people at UNCC). Now, I'm trying to get Xilinx SystemACE device driver
>> support and Ethernet working. Unfortunately, I do not know much about
>> adding support for a driver that does not currently exist in the kernel.
>> Does anybody know if it would be possible to get SysACE driver support
>> in 2.6.10? Downloading an ELF file onto a board using a JTAG cable is
>> slow and inconvenient...
>>
>> My attempt at adding SysACE driver support involved adding in
>> drivers/block/xsysace.c and linux/platform_device.h (files pulled from
>> mainline Linux 2.6.24) and editing drivers/block/Kconfig and
>> drivers/block/Makefile. The kernel failed to compile, however, and spit
>> out these errors:
>>
>>
>> In file included from drivers/block/xsysace.c:93:
>> include/linux/platform_device.h:16: error: redefinition of `struct
>> platform_device'
>> include/linux/platform_device.h:39: error: conflicting types for
>> 'platform_device_register_simple'
>> include/linux/device.h:387: error: previous declaration of
>> 'platform_device_register_simple' was here
>> include/linux/platform_device.h:39: error: conflicting types for
>> 'platform_device_register_simple'
>> include/linux/device.h:387: error: previous declaration of
>> 'platform_device_register_simple' was here
>> include/linux/platform_device.h:52: error: syntax error before
>> "pm_message_t"
>> include/linux/platform_device.h:52: warning: function declaration isn't
>> a prototype
>> include/linux/platform_device.h:53: error: syntax error before
>> "pm_message_t"
>> include/linux/platform_device.h:53: warning: function declaration isn't
>> a prototype
>> drivers/block/xsysace.c: In function `ace_fsm_yieldirq':
>> drivers/block/xsysace.c:461: error: `NO_IRQ' undeclared (first use in
>> this function)
>> drivers/block/xsysace.c:461: error: (Each undeclared identifier is
>> reported only once
>> drivers/block/xsysace.c:461: error: for each function it appears in.)
>> drivers/block/xsysace.c: In function `ace_fsm_dostate':
>> drivers/block/xsysace.c:723: error: too many arguments to function
>> `end_that_request_last'
>> drivers/block/xsysace.c: At top level:
>> drivers/block/xsysace.c:931: error: unknown field `getgeo' specified in
>> initializer
>> drivers/block/xsysace.c:931: warning: initialization from incompatible
>> pointer type
>> drivers/block/xsysace.c: In function `ace_setup':
>> drivers/block/xsysace.c:960: warning: implicit declaration of function
>> `setup_timer'
>> drivers/block/xsysace.c:1009: error: `NO_IRQ' undeclared (first use in
>> this function)
>> drivers/block/xsysace.c:1010: warning: passing arg 2 of `request_irq'
>> from incompatible pointer type
>> drivers/block/xsysace.c: In function `ace_teardown':
>> drivers/block/xsysace.c:1061: error: `NO_IRQ' undeclared (first use in
>> this function)
>> drivers/block/xsysace.c: In function `ace_alloc':
>> drivers/block/xsysace.c:1081: warning: implicit declaration of function
>> `kzalloc'
>> drivers/block/xsysace.c:1081: warning: assignment makes pointer from
>> integer without a cast
>> drivers/block/xsysace.c: In function `ace_probe':
>> drivers/block/xsysace.c:1131: error: `NO_IRQ' undeclared (first use in
>> this function)
>> make[2]: *** [drivers/block/xsysace.o] Error 1
>> make[1]: *** [drivers/block] Error 2
>> make: *** [drivers] Error 2
>>
>>
>> Also, is anybody familiar with the Ethernet drivers? My boot-log
>> contains this error concerning Ethernet:
>>
>>
>> eepro100.c:v1.09j-t 9/29/99 Donald Becker
>> http://www.scyld.com/network/eepro100.html
>> eepro100.c: $Revision: 1.36 $ 2000/11/17 Modified by Andrey V. Savochkin
>> <saw at saw.sw.com.sg> and others
>> eth0: Invalid EEPROM checksum 0xbf8a, check settings before activating
>> this device!
>> eth0: OEM i82557/i82558 10/100 Ethernet, 00:0A:35:00:7C:A0, IRQ 0.
>>  Board assembly 727095-004, Physical connectors present: RJ45
>>  Primary interface chip i82555 PHY #1.
>> Self test failed, status ffffffff:
>> Failure to initialize the i82557.
>> Verify that the card is a bus-master capable slot.
>>
>>
>> I also tried another Ethernet driver and got this error instead:
>>
>>
>> e100: Intel(R) PRO/100 Network Driver, 3.2.3-k2-NAPI
>> e100: Copyright(c) 1999-2004 Intel Corporation
>> e100: 0000:00:07.0: e100_eeprom_load: EEPROM corrupted
>> e100: probe of 0000:00:07.0 failed with error -11
>>
>>
>> I know that newer versions of mainline and Xilinx Linux 2.6 have SysACE
>> support, but I would prefer to stick with mainline 2.6.10 because, well,
>> it sort of works for me at the moment (I was able to mount an ATA hard
>> drive with an ext2 file system). Any help/suggestions would be greatly
>> appreciated. Let me know if you need me to provide any additional
>> information. Thanks!
>>
>>
>> -Juliana
>>
>>
>>
>>
>>
>> ------------------------------
>>
>> _______________________________________________
>> Linuxppc-embedded mailing list
>> Linuxppc-embedded at ozlabs.org
>> https://ozlabs.org/mailman/listinfo/linuxppc-embedded
>>
>> End of Linuxppc-embedded Digest, Vol 50, Issue 8
>> ************************************************
>>
>
>
>
> ------------------------------
>
> Message: 2
> Date: Wed, 8 Oct 2008 15:51:35 -0700
> From: "Duy-Ky Nguyen" <duykynguyen at hotmail.com>
> Subject: Failure in re-load module
> To: <linuxppc-embedded at ozlabs.org>
> Cc: v.selvamuthukumar at gmail.com
> Message-ID: <BLU119-DAV1197362B77CCC3F1296F60BB3B0 at phx.gbl>
> Content-Type: text/plain; format=flowed; charset="iso-8859-1";
> reply-type=original
>
> Hello,
>
> I'm able to load driver and access HW without any problem
> In removing the driver using command "rmmod", I have error relating to
> virtual address from ioremap() for physical adress IMMR_BASE = 0xE0000000
>
> Trying to free nonexistent resource <00000000c98a0000-00000000c98affff>
>
> But the driver appears removed when I check with command "lsmod"
> However, I'm NOT able to load the driver again due to error
>
> PPC: can't get I/O port address 0xe0000000
> insmod: cannot insert `ppc_drv.ko': No such device (-1): No such device
>
> It appaers to me that the memory region was not lreleased successfully
>
> My questions are
> 1) What is the right sequence in requesting and releasing memory region ?
> 2) Are the function request_region() and release_region() obsolete in 
> kernel
> 2.6 as I have no such problem with kernel 2.4 ?
>
> I have a driver with basic code segment below
> ///////////////////////////// Start of code
> //////////////////////////////////// Init
> init:
> if (! request_region(PPC_BASE, PPC_SIZE, DEVICE_NAME)) {
>  printk(KERN_INFO "PPC: can't get I/O port address 0x%lx\n", ppc_base);
>  return -ENODEV;
> }
>
> ppc_base = (unsigned long) ioremap(PPC_BASE, PPC_SIZE);
>
> result = register_chrdev(major, DEVICE_NAME, &ppc_fops);
>
> //////////////////////////// exit
> exit:
> unregister_chrdev(major, DEVICE_NAME);
>
> iounmap((void __iomem *)ppc_base);
> iounmap((void __iomem *)lbc_base);
>
> release_region(ppc_base,PPC_SIZE);
> release_region(lbc_base,PPC_SIZE);
> ////////////////////////////// End of code
>
> Thanks so much for your time and help.
>
> Best Regards,
>
> Duy-Ky
>
>
>
> ------------------------------
>
> Message: 3
> Date: Wed, 08 Oct 2008 17:54:41 -0500
> From: Scott Wood <scottwood at freescale.com>
> Subject: Re: Failure in re-load module
> To: Duy-Ky Nguyen <duykynguyen at hotmail.com>
> Cc: v.selvamuthukumar at gmail.com, linuxppc-embedded at ozlabs.org
> Message-ID: <48ED3A31.60503 at freescale.com>
> Content-Type: text/plain; charset=UTF-8; format=flowed
>
> Duy-Ky Nguyen wrote:
>> release_region(ppc_base,PPC_SIZE);
>> release_region(lbc_base,PPC_SIZE);
>
> This should be PPC_BASE/LBC_BASE (the physical address), not
> ppc_base/lbc_base (the virtual address).
>
> -Scott
>
>
> ------------------------------
>
> _______________________________________________
> Linuxppc-embedded mailing list
> Linuxppc-embedded at ozlabs.org
> https://ozlabs.org/mailman/listinfo/linuxppc-embedded
>
> End of Linuxppc-embedded Digest, Vol 50, Issue 12
> *************************************************
>