January 2005 - U-Boot

[U-Boot-Users] Question on do_bootm in cmd_bootm.c
by Nick Barendt 31 Mar '05

31 Mar '05

Can someone, maybe Wolfgang, comment on the maximum 4MB uncompressed size of gunzip (and, possibly, bzip2) images? I've been working on a Coldfire 5272 system for quite some time with u-boot and uClinux. Both have been remarkably stable. We use the uClinux default build wherein the root filesystem (romfs) is appended to the end of the kernel image, starting where the bss would normally be. The kernel startup code relocates the romfs image to just beyond the bss, then, and everything runs smoothly. I know that many folks in the PPC world use separate kernel and root filesystem images, for many good reasons, but many (most?) of the uClinux platforms use this concatenated kernel+romfs image. Recently, we've been adding more content to the root filesystem. As our root filesystem grew larger (say about 3MB), however, we started seeing boot problems where the kernel would boot fine but the root filesystem was clearly corrupt. I've spent the better part of two days trying to track down where things go wrong, from the kernel to u-boot. Of course, I used a hardware debugger, looked at u-boot's memory assignments, image load addresses, entry points, malloc areas, etc. I just located the source of my trouble, a hard-coded 4MB maximum uncompressed size in do_bootm: int do_bootm (cmd_tbl_t *cmdtp, int flag, int argc, char *argv[]) { ulong iflag; ulong addr; ulong data, len, checksum; ulong *len_ptr; uint unc_len = 0x400000; ^^^^^^^^^ int i, verify; SNIP case IH_COMP_GZIP: printf (" Uncompressing %s ... ", name); if (gunzip ((void *)ntohl(hdr->ih_load), unc_len, ^^^^^^^^ (uchar *)data, (int *)&len) != 0) { printf ("GUNZIP ERROR - must RESET board to recover\n"); SHOW_BOOT_PROGRESS (-6); do_reset (cmdtp, flag, argc, argv); } break; SNIP Increasing unc_len to larger values (like 6MB) seems to solve my problem. With an uncompressed image larger than 4MB, gunzip doesn't seem to generate any errors (granted, I haven't stepped through gunzip and zlib), but merrily runs along, although the data beyond the 4MB boundary seems to be missing or corrupt. It looks like there's a truncation (about line 1875 in zlib.c) to z.avail_out in zlib that prevents data from being copied beyond unc_len - maybe that should be an error instead of silently truncating data? Just a thought. I'll into submitting a patch for that. I was wondering if there was some reason for the 0x400000 magic number for unc_len. Is that just a magic number? Is that larger than anyone uses for images? I guess I'm a little surprised that this hasn't bitten anyone else so far (nothing I could see on the mailing list archive, at least). Our 1MB kernel and 3MB romfs root filesystem might be exceptionally large, I don't know. Would there be any issues with making unc_len a CFG_ macro, if I submitted a patch? Nick Barendt

4 8

[U-Boot-Users] PATCH: Nios-II cleanups
by Scott McNutt 31 Mar '05

31 Mar '05

Patch has some minor Nios-II cleanups: -Add sysid command (Nios-II only). -Locate default exception trampoline at proper offset. -Implement I/O routines (readb, writeb, etc). -Implement do_bootm_linux Regards, --Scott diff -Naurb u-boot-2004-10-20/cpu/nios2/Makefile u-boot-work/cpu/nios2/Makefile --- u-boot-2004-10-20/cpu/nios2/Makefile 2004-10-10 17:27:33.000000000 -0400 +++ u-boot-work/cpu/nios2/Makefile 2004-10-11 21:38:41.000000000 -0400 @@ -27,7 +27,7 @@ START = start.o AOBJS = exceptions.o -OBJS = cpu.o interrupts.o serial.o traps.o +OBJS = cpu.o interrupts.o serial.o sysid.o traps.o all: .depend $(START) $(LIB) diff -Naurb u-boot-2004-10-20/cpu/nios2/cpu.c u-boot-work/cpu/nios2/cpu.c --- u-boot-2004-10-20/cpu/nios2/cpu.c 2004-10-10 17:27:33.000000000 -0400 +++ u-boot-work/cpu/nios2/cpu.c 2004-10-11 21:40:55.000000000 -0400 @@ -26,19 +26,7 @@ #include <nios2-io.h> #if defined (CFG_NIOS_SYSID_BASE) -#include <linux/time.h> -static void check_sysid (void) -{ - struct nios_sysid_t *sysid = - (struct nios_sysid_t *)CACHE_BYPASS(CFG_NIOS_SYSID_BASE); - struct tm t; - char asc[32]; - - localtime_r ((time_t *)&sysid->timestamp, &t); - asctime_r (&t, asc); - printf ("SYSID : %08x, %s", sysid->id, asc); - -} +extern void display_sysid (void); #endif /* CFG_NIOS_SYSID_BASE */ int checkcpu (void) @@ -47,7 +35,7 @@ #if !defined(CFG_NIOS_SYSID_BASE) printf ("SYSID : <unknown>\n"); #else - check_sysid (); + display_sysid (); #endif return (0); } diff -Naurb u-boot-2004-10-20/cpu/nios2/start.S u-boot-work/cpu/nios2/start.S --- u-boot-2004-10-20/cpu/nios2/start.S 2004-10-10 17:27:33.000000000 -0400 +++ u-boot-work/cpu/nios2/start.S 2004-10-20 10:50:47.000000000 -0400 @@ -39,14 +39,24 @@ * just be invalidating the cache a second time. If cache * is not implemented initi behaves as nop. */ - movhi r4, %hi(CFG_ICACHELINE_SIZE) - ori r4, r4, %lo(CFG_ICACHELINE_SIZE) + ori r4, r0, %lo(CFG_ICACHELINE_SIZE) movhi r5, %hi(CFG_ICACHE_SIZE) ori r5, r5, %lo(CFG_ICACHE_SIZE) mov r6, r0 0: initi r6 add r6, r6, r4 bltu r6, r5, 0b + br _except_end /* Skip the tramp */ + + /* EXCEPTION TRAMPOLINE -- the following gets copied + * to the exception address (below), but is otherwise at the + * default exception vector offset (0x0020). + */ +_except_start: + movhi et, %hi(_exception) + ori et, et, %lo(_exception) + jmp et +_except_end: /* INTERRUPTS -- for now, all interrupts masked and globally * disabled. @@ -117,7 +127,9 @@ _reloc: /* COPY EXCEPTION TRAMPOLINE -- copy the tramp to the - * exception address. + * exception address. Define CONFIG_ROM_STUBS to prevent + * the copy (e.g. exception in flash or in other + * softare/firmware component). */ #if !defined(CONFIG_ROM_STUBS) movhi r4, %hi(_except_start) @@ -126,12 +138,14 @@ ori r5, r5, %lo(_except_end) movhi r6, %hi(CFG_EXCEPTION_ADDR) ori r6, r6, %lo(CFG_EXCEPTION_ADDR) + beq r4, r6, 7f /* Skip if at proper addr */ 6: ldwio r7, 0(r4) stwio r7, 0(r6) addi r4, r4, 4 addi r6, r6, 4 bne r4, r5, 6b +7: #endif /* STACK INIT -- zero top two words for call back chain. @@ -155,15 +169,6 @@ */ br _start - /* EXCEPTION TRAMPOLINE -- the following gets copied - * to the exception address. - */ -_except_start: - movhi et, %hi(_exception) - ori et, et, %lo(_exception) - jmp et -_except_end: - /* * dly_clks -- Nios2 (like Nios1) doesn't have a timebase in diff -Naurb u-boot-2004-10-20/cpu/nios2/sysid.c u-boot-work/cpu/nios2/sysid.c --- u-boot-2004-10-20/cpu/nios2/sysid.c 1969-12-31 19:00:00.000000000 -0500 +++ u-boot-work/cpu/nios2/sysid.c 2004-10-11 21:45:42.000000000 -0400 @@ -0,0 +1,57 @@ +/* + * (C) Copyright 2004, Psyent Corporation <www.psyent.com> + * Scott McNutt <smcnutt(a)psyent.com> + * + * See file CREDITS for list of people who contributed to this + * project. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, + * MA 02111-1307 USA + */ + +#include <common.h> + +#if defined (CFG_NIOS_SYSID_BASE) + +#include <command.h> +#include <nios2.h> +#include <nios2-io.h> +#include <linux/time.h> + +void display_sysid (void) +{ + struct nios_sysid_t *sysid = + (struct nios_sysid_t *)CACHE_BYPASS(CFG_NIOS_SYSID_BASE); + struct tm t; + char asc[32]; + + localtime_r ((time_t *)&sysid->timestamp, &t); + asctime_r (&t, asc); + printf ("SYSID : %08x, %s", sysid->id, asc); + +} + +int do_sysid (cmd_tbl_t *cmdtp, int flag, int argc, char *argv[]) +{ + display_sysid (); + return (0); +} + +U_BOOT_CMD( + sysid, 1, 1, do_sysid, + "sysid - display Nios-II system id\n\n", + "\n - display Nios-II system id\n" +); +#endif /* CFG_NIOS_SYSID_BASE */ diff -Naurb u-boot-2004-10-20/include/asm-nios2/io.h u-boot-work/include/asm-nios2/io.h --- u-boot-2004-10-20/include/asm-nios2/io.h 2004-10-10 17:27:34.000000000 -0400 +++ u-boot-work/include/asm-nios2/io.h 2004-10-18 23:57:39.000000000 -0400 @@ -29,8 +29,61 @@ extern unsigned char inb (unsigned char *port); extern unsigned short inw (unsigned short *port); extern unsigned inl (unsigned port); -extern void outb (unsigned char val, unsigned char *port); -extern void outw (unsigned short val, unsigned short *port); -extern void outl (unsigned val, unsigned port); + +#define readb(addr)\ + ({unsigned char val;\ + asm volatile( "ldbio %0, 0(%1)" :"=r"(val) : "r" (addr)); val;}) +#define readw(addr)\ + ({unsigned short val;\ + asm volatile( "ldhio %0, 0(%1)" :"=r"(val) : "r" (addr)); val;}) +#define readl(addr)\ + ({unsigned long val;\ + asm volatile( "ldwio %0, 0(%1)" :"=r"(val) : "r" (addr)); val;}) +#define writeb(addr,val)\ + asm volatile ("stbio %0, 0(%1)" : : "r" (addr), "r" (val)) +#define writew(addr,val)\ + asm volatile ("sthio %0, 0(%1)" : : "r" (addr), "r" (val)) +#define writel(addr,val)\ + asm volatile ("stwio %0, 0(%1)" : : "r" (addr), "r" (val)) + +#define inb(addr) readb(addr) +#define inw(addr) readw(addr) +#define inl(addr) readl(addr) +#define outb(addr,val) writeb(addr,val) +#define outw(addr,val) writew(addr,val) +#define outl(addr,val) writel(addr,val) + +static inline void insb (unsigned long port, void *dst, unsigned long count) +{ + unsigned char *p = dst; + while (count--) *p++ = inb (port); +} +static inline void insw (unsigned long port, void *dst, unsigned long count) +{ + unsigned short *p = dst; + while (count--) *p++ = inw (port); +} +static inline void insl (unsigned long port, void *dst, unsigned long count) +{ + unsigned long *p = dst; + while (count--) *p++ = inl (port); +} + +static inline void outsb (unsigned long port, const void *src, unsigned long count) +{ + const unsigned char *p = src; + while (count--) outb (*p++, port); +} + +static inline void outsw (unsigned long port, const void *src, unsigned long count) +{ + const unsigned short *p = src; + while (count--) outw (*p++, port); +} +static inline void outsl (unsigned long port, const void *src, unsigned long count) +{ + const unsigned long *p = src; + while (count--) outl (*p++, port); +} #endif /* __ASM_NIOS2_IO_H_ */ diff -Naurb u-boot-2004-10-20/lib_nios2/nios_linux.c u-boot-work/lib_nios2/nios_linux.c --- u-boot-2004-10-20/lib_nios2/nios_linux.c 2004-10-10 17:27:35.000000000 -0400 +++ u-boot-work/lib_nios2/nios_linux.c 2004-10-20 10:29:35.000000000 -0400 @@ -23,10 +23,18 @@ #include <common.h> #include <command.h> +#include <asm/byteorder.h> + +extern image_header_t header; /* common/cmd_bootm.c */ -/* TODO - */ void do_bootm_linux(cmd_tbl_t *cmdtp, int flag, int argc, char *argv[], ulong addr, ulong *len_ptr, int verify) { + image_header_t *hdr = &header; + void (*kernel)(void) = (void (*)(void))ntohl (hdr->ih_ep); + + /* For now we assume the Microtronix linux ... which only + * needs to be called ;-) + */ + kernel (); }

2 1

[U-Boot-Users] Make drivers/cfi_flash.c protect environment and redundant enviro nment
by Michael Bendzick 26 Feb '05

26 Feb '05

Included is a patch to the drivers/cfi_flash.c driver that will automatically enable environment and redundant environment protection protection upon booting. Previously, only the monitor (U-Boot) was protected by the driver. Change log snippet: * Patch by Michael Bendzick, 12 Jul 2004: - Make drivers/cfi_flash.c protect environment and redundant environment. -Michael Bendzick Systems and Software Engineering Logic Product Development michael.b(a)logicpd.com 612-436-5122 www.logicpd.com

3 2

[U-Boot-Users] JFFS2 images written from U-Boot unusable in Linux? (Round 2)
by Martin Egholm Nielsen 16 Feb '05

16 Feb '05

Hi there, Now I've gotten a bit further in my quest for getting JFFS2 images to work between U-Boot and Linux. I upgraded to ELDK 3.1 and configured the CONFIG_JFFS2_NAND_SIZE to the proper size. So, now I can succesfully put an image on the flash from U-Boot and "ls" it. (Thanks again for that...) Further, if I erase the NAND from Linux and create some files/dir's, I can see them succesfully from U-Boot with "ls". However, if I put a JFFS2-image on the NAND _from U-Boot_, Linux cannot mount the device any longer: # cd /tmp # mkdir image # cd image/ # echo "Test1" > testfile # mkdir testdir # echo "Test12" > testdir/testfile # mkfs.jffs2 --pad --big-endian --root=. --output=../test20050107.jffs2 => nand erase clean => tftp 100000 test20050107.jffs2 => nand write.jffs2 100000 0 $(filesize) => ls Scanning JFFS2 FS: . done. drwxr-xr-x 0 Tue Nov 30 17:45:59 2004 testdir -rw-r--r-- 6 Tue Nov 30 17:45:45 2004 testfile => ls testdir -rw-r--r-- 7 Tue Nov 30 17:45:59 2004 testfile # mkdir /mnt # mount -t jffs2 /dev/mtdblock0 /mnt jffs2: Erase block size too small (16KiB). Using virtual blocks size (32KiB) instead Cowardly refusing to erase blocks on filesystem with no valid JFFS2 nodes empty_blocks 0, bad_blocks 5, c->nr_blocks 2048 mount: Mounting /dev/mtdblock0 on /mnt failed: Invalid argument I have the following kernel configuration options concerning JFFS2: CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 CONFIG_JFFS2_FS_NAND=y CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y # CONFIG_JFFS2_RUBIN is not set # CONFIG_JFFS2_LZO is not set # CONFIG_JFFS2_LZARI is not set # CONFIG_JFFS2_CMODE_NONE is not set CONFIG_JFFS2_CMODE_PRIORITY=y # CONFIG_JFFS2_CMODE_SIZE is not set # CONFIG_JFFS2_PROC is not set Are there anything else needed in order to support the way U-Boot writes an image? Best regards, Martin Egholm

5 9

[U-Boot-Users] Environment in flash rather than EEProm
by Andy Hawkins 11 Feb '05

11 Feb '05

Hi all I'm using a PPChameleonEVB, and am trying to move the u-boot environment into flash. I've modified the include/configs/PPChameleon.h file as follows: #define CFG_ENV_IS_IN_FLASH 1 #define CFG_ENV_ADDR 0xFFFFA000 #define CFG_ENV_OFFSET (CFG_ENV_ADDR - 0xFFFC0000) #define CFG_ENV_SIZE 0x2000 I've added some debugging into the section where it reads the environment, and I'm seeing the following: U-Boot 1.0.2-adh-2004/03/29 (Mar 30 2004 - 13:40:38) CPU: IBM PowerPC 405EP Rev. B at 133.333 MHz (PLB=133, OPB=66, EBC=33 MHz) IIC Boot EEPROM disabled PCI async ext clock used, internal PCI arbiter enabled 16 kB I-Cache 16 kB D-Cache Board: ### No HW ID - assuming PPChameleonEVB I2C: ready DRAM: 32 MB Top of RAM usable for U-Boot at: 02000000 Reserving 189k for U-Boot at: 01fd0000 Reserving 256k for malloc() at: 01f90000 Reserving 128 Bytes for Board Info at: 01f8ff80 Reserving 48 Bytes for Global Data at: 01f8ff50 Stack Pointer at: 01f8ff38 New Stack Pointer is: 01f8ff38 Now running in RAM - U-Boot at: 01fd0000 FLASH: 4 MB env_relocate[204] offset = 0x2010000 env_relocate[216] embedded ENV (before reloc) at fffd9474 env_relocate[218] embedded ENV (after reloc) at 01fe9474 U-Boot relocated to 01fd0000 NAND:Probing at 0xff000000 32 MB ### main_loop entered: bootdelay=5 ### main_loop: bootcmd="<UNDEFINED>" => As you can see, the address of the environment before relocation is 0xfffd9474. Shouldn't this be 0xFFFFA000? Is there anything else I need to do to force the location of the environment into the required place? It looks like the address of the environment is based on the location of the 'environment' variable in common/environment.c. Do I need to change the LD script to force this variable to a certain location? If so, can anyone help with what I need to do (I'm not familiar with LD scripts yet). Many thanks. Andy

3 7

[U-Boot-Users] duplicate environment variables in U-Boot? (full log)
by Clemens Koller 04 Feb '05

04 Feb '05

Hello, again... I forgot to attach the full log to help you to get an idea of what I mean. Some of the 'Unknown commands' should be known, actually. Best regards, Clemens Koller _______________________________ R&D Imaging Devices Anagramm GmbH Rupert-Mayer-Str. 45/1 81379 Muenchen Germany http://www.anagramm.de Phone: +49-89-741518-50 Fax: +49-89-741518-19

3 4

[U-Boot-Users] fs/ext2/ext2fs.c bug and patch
by Andrew Dyer 04 Feb '05

04 Feb '05

Hi all, In the course of getting ext2 support to work on our platform I noticed that the ext2load command calls ext2fs_open() and uses a return code of less than 0 as an error indication, otherwise the return value is a length. ext2fs_open() uses 0 to indicate a returned error, so the ext2load command doesn't properly flag errors coming back from the filesystem code. The ext2load command is the only caller of ext2fs_open() that I could see. The attached patch changes the return codes from ext2fs_open() to be negative on an error. -- Hardware, n.: The parts of a computer system that can be kicked.

2 1

[U-Boot-Users] Re: Re: really really tiny code
by Nishant Galange 02 Feb '05

02 Feb '05

Sorry about the off-topic email. I sincerely felt that the people on this mailing list are the only ones who can help me. About the problem itself: Yes, I am a little comfortable with the u-boot source code now, and I planned on using most of the code in cpu/mpc8260/start.S for making this 'micro monitor' I definitely plan to make it available under the GPL, as I feel it will be a really good test tool for people bringing up their new hardware for the first time. They can use the processor to test crucial parts of the system, and use LEDs to report status. From my (limited) experience in board bring-up, everything right from the power supply to bad buffers, JTAG connections to bus noise are suspect when the board dosen't work. Atleast this tiny code will help clear up doubts from the processor's point of view. Developers will have one definite ally on the board who keeps sending status reports no matter what else is failing. The only help I need is with gcc, or ppc_82xx-gcc, to be precise. I have no idea how to give it a single .S or .asm file and tell it to create a .bin. Makefiles? I only know how to use readymade ones, not to how to create one. You said I will also need a .lds, and I don't know how to tell gcc to use that either. Can anyone please send the complete process of compiling such a source (only 1 or two source files) along with the .lds and creating a .bin? Waiting, Nishant PS. Sorry again if its still off-topic. But I feel it can be a good addition to u-boot as a separate tool. This will also be independent of the board, and most people will simply have to change the addresses of the LEDs to get it working for their board. -----Original Message----- From: wd(a)denx.de [mailto:wd@denx.de] Sent: Wednesday, January 26, 2005 3:05 PM To: Nishant Galange Cc: u-boot-users(a)lists.sourceforge.net Subject: Re: [U-Boot-Users] really really tiny code In message <1106684792.3804b95cnishant.subs(a)myrealbox.com> you wrote: > > I want to write a really tiny program - just init core, MMU and stop. > Preferably, (if possible) I would like to start out with a blank file, > type in the code, and give the compile command at the good old command prom= > pt. > > Is it possible? If yes, can anyone please tell me the command? Yes, it is poossible. You will need more than one command, at least "vi" and "ppc_82xx-gcc". > Would any other files (.lds, etc) be necessary? What else will be required? Yes, you will most probably need a linker script. One of the most important requirements is an in-depth understanding of your hardware environment and how to write software for it. > I need a micro development environment like this from where I can do all > my work out of the cosy little (really limited, I know) DPRAM, and try > out some experiments with the processor's registers and config. > I might even be adventurous and try initializing the ports and > glowing some LEDs :-) > > What do I do? How do I proceed? Please help. U-Boot includes all the init stuff you will probably re-0invent, so feel free to copy and re-use this code (and make your program available under GPL), but please understand that you are on your own with such a project. Sorry, but this is off topic here. Best regards, Wolfgang Denk -- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd(a)denx.de A year spent in artificial intelligence is enough to make one believe in God.

3 2

[U-Boot-Users] do_run_[os]. why restart if there is a minor problem?
by Ara Avanesyan 01 Feb '05

01 Feb '05

Hi, in function do_run_linux if it finds ramdisk's crc is bad or it's not "u-boot-ized" (mkimage) it resets the board. and many other similar minor things. so why restart? I propose making do_run_linux return something to indicate there is no need to reboot the board and bootm checks this value and returns. if you find this may cause backwards compatibility problems, let's make it depending on some preprocessor define. I can submit a patch if you say ok. __ Thanks, Ara

2 4

[U-Boot-Users] new port/code location
by Marc Leeman 01 Feb '05

01 Feb '05

We have a port of a PPC 8245 based board and I would want to send a patch upstream with the board support files. I still need to port it to the 1.1.2 release in CVS, but I have one main question. At the end of the initialisation (before the main_loop()), we shortcut to our own boot sequence barcbcd_boot(); The boot code reads a number of bytes from a location from flash, and based on the values of these bytes, we boot the kernel locate 0xA or 0xB (after updating the value in flash). The practical effect is: try this kernel (and configuration); if it fails, an external source resets the PPC and we'll boot a version we know will work the next time around. The only code changed in lib_ppc/board.c are: #ifdef CONFIG_BARCOBCD_STREAMING barcobcd_boot(); #endif The question is if this is acceptable and if not, what would be a better location to obtain the same effect. -- greetz, marc I'm Butch. This is Sundance. Crichton - Til the Blood Runs Clear scorpius 2.6.10 #1 Mon Dec 27 13:05:02 CET 2004 GNU/Linux

2 2