ltib
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Ltib] Kernel panic after update the latest source for LPC32xx targe


From: Kevin Wells
Subject: RE: [Ltib] Kernel panic after update the latest source for LPC32xx target
Date: Wed, 4 Aug 2010 19:39:39 +0200

 

Hi Henry,

 

There is a website going live in a few days that will track patch history better.

This won’t help you yet, so here are the latest updates to 2.6.37.8 and 2.6.34. You can also browse the CVS history of the ./ltid/dist/lfs-5.1/kernel/kernel-xxxx-lpc32xx.spec.in files to get patch history and view the raw patches in the /opt/ltib/pkgs directory.

 

Kernel version 2.6.27.8 updated 07/19/2010

  • Updated default kernel configs for PHY3250 and EA3250 boards
  • Added suspend to mem power management support
  • Fixed an issue with USB clock/enable disable that could sometimes prevent USB clock from starting
  • Fixed scatter/gather linked list support (fixes an audio issue)
  • Added support in DMA driver for scatter/gather specific for NAND and software burst enable
  • Re-sized and added/removed MTD partitions for NAND for EA and Phytec boards
  • Added DMA support for NAND/MTD for EA and Phytec boards
  • Removed wakeup event clear logic from IRQ driver when interrupt is cleared, not needed
  • Added a simple GPIO interrupt and event wakeup example for the Phytec board using the BTN1 button
  • Added wakeup support to the touchscreen driver
  • Added suspend/resume functions to the high speed UART driver
  • NAND/MTD driver now supports hardware ECC
  • Fixed and improved alarm functions in the RTC driver
  • Fixed and enabled gadget driver DMA support
  • Fixed an issue with the gadget driver where the USB D pulldown would get stuck asserted when the module was removed
  • Fixed several clock crash issues with the gadget driver related to module removal and cable removal

Kernel version 2.6.34 updated 07/16/2010

  • Fixed an issue with USB clock/enable disable that could sometimes prevent USB clock from starting
  • Revised SD card clock rate set, get, and rounding functions
  • Fixed scatter/gather linked list support (fixes an audio issue)
  • Added support in DMA driver for scatter/gather specific for NAND and software burst enable
  • Fixed a potential compilation issue in the ea3250.c file when the LCD is disabled
  • Re-sized and added/removed MTD partitions for NAND for EA and Phytec boards
  • Added DMA support for NAND/MTD for EA and Phytec boards
  • Fixed an issue with gpiolib that would return a wrong state for GPIO banks 3 input pins
  • Fixed several incorrectly named definitions (was LCP, now LPC)
  • Removed wakeup event clear logic from IRQ driver when interrupt is cleared, not needed
  • SD/MMC driver data handling completely re-written, now supports scatter/gather and is better optimized
  • Updated card detect, write protect, and power control logic for the SD card controller driver for the Phytec board (uses driver functions now instead of callbacks)
  • Added a simple GPIO interrupt and event wakeup example for the Phytec board using the BTN1 button
  • Added wakeup support to the touchscreen driver
  • Added suspend/resume functions to the high speed UART driver
  • NAND/MTD driver now supports hardware ECC
  • Fixed and improved alarm functions in the RTC driver
  • Fixed and enabled gadget driver DMA support
  • Fixed an issue with the gadget driver where the USB D pulldown would get stuck asserted when the module was removed
  • Fixed several clock crash issues with the gadget driver related to module removal and cable removal
  • Added default configurations in arch/arm/configs for the EA3250 and PHY3250 boards

And here is the u-boot 2009.03 changes updated 07/16/2010

  • Moved NAND driver to common NAND area, removed Phytec and EA NAND drivers
  • NAND driver now support hardware ECC
  • Added basic DMA driver for NAND support
  • Revised u-boot environment save locations and sizes (matches S1L? and kernel)

thanks,

Kevin

 

 

From: address@hidden [mailto:address@hidden On Behalf Of Li Yuan-Lung
Sent: Tuesday, August 03, 2010 7:58 PM
To: Kevin Wells
Cc: address@hidden
Subject: Re: [Ltib] Kernel panic after update the latest source for LPC32xx target

 

Dear Kevin,

Thanks again, I want to know where can I find a document for us to follow on the latest patch easier. For example, the change log in somewhere or not?

 

Henry

2010/8/4 Kevin Wells <address@hidden>

 

Hi Henry,

 

This was changed to fix an issue with potential overwrite of S1L application code (via nsave, nload) with the u-boot parameters.

If this occurred, the board wouldn’t boot u-boot anymore.

An additional u-boot partition (/dev/mtd1) was also added so u-boot could be updated from the kernel.

 

I don’t suspect this mapping to change again in the LTIB patch series.

Sorry, I hope this didn’t cause too many issues. These types of changes are rare, but sometimes they are needed.

 

thanks,

Kevin

 

 

Dear all,

Finally I found where the problem is.

The mem-map changed in MTD block!

The old one is:

Creating 4 MTD partitions on "lpc32xx_nand":

0x000000000000-0x000000168000 : "phy3250-boot"

0x000000168000-0x000000190000 : "phy3250-ubt-prms"

0x000000190000-0x000000590000 : "phy3250-kernel"

0x000000590000-0x000004000000 : "phy3250-rootfs"

 

The new one is:

Creating 5 MTD partitions on "lpc32xx_nand":

0x000000000000-0x000000064000 : "phy3250-boot"

0x000000064000-0x0000001f4000 : "phy3250-uboot"

0x0000001f4000-0x000000204000 : "phy3250-ubt-prms"

0x000000204000-0x000000604000 : "phy3250-kernel"

0x000000604000-0x000004000000 : "phy3250-rootfs"

 

Could anybody or any document tell me why? Will this mapping be changed again in the future?

This changes impact the burning procedure of our sample target!

 

Henry

 

2010/8/3 Li Yuan-Lung <henry.li@climax.com.tw>

Dear NXP friends,

After using "ltib -m distclean" and re-build the kernel, the booting becomes kernel panic, the build config is just the same as old one. Is there any modification need to be change with the menuconfig? Thank you.

 

The panics messages:

   ........................................................................................

jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00188024: 0x7366 instead

Further such events for this erase block will not be printed

VFS: Mounted root (jffs2 filesystem) on device 31:3.

Freeing init memory: 116K

Failed to execute /sbin/init.  Attempting defaults...

Kernel panic - not syncing: No init found.  Try passing init= option to kernel. See Linux Documentation/init.txt for guidance.

Backtrace:

[<c00299fc>] (dump_backtrace+0x0/0x108) from [<c0029f58>] (dump_stack+0x18/0x1c)

 r6:00000000 r5:00000000 r4:c0359834

[<c0029f40>] (dump_stack+0x0/0x1c) from [<c00440f0>] (panic+0x5c/0xf8)

[<c0044094>] (panic+0x0/0xf8) from [<c00255d8>] (init_post+0xb8/0xe8)

 r3:c3c05e08 r2:00000003 r1:00000002 r0:c02ed374

[<c0025520>] (init_post+0x0/0xe8) from [<c0008998>] (kernel_init+0x120/0x160)

 r4:c0359838

[<c0008878>] (kernel_init+0x0/0x160) from [<c0047598>] (do_exit+0x0/0x684)

 r4:00000000

 

 

Henry




--
Best Regards,

Henry Li
SW Engineer of Climax Co.
TEL: +886-2-2794-0001 ext.231
FAX: +886-2-2792-6618




--
Best Regards,

Henry Li
SW Engineer of Climax Co.
TEL: +886-2-2794-0001 ext.231
FAX: +886-2-2792-6618


reply via email to

[Prev in Thread] Current Thread [Next in Thread]