This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
RE: Probleme with ram rom data sections
- From: Matthieu dot GIRARDIN at fr dot thalesgroup dot com
- To: gary at mlbassoc dot com
- Cc: ecos-discuss at sources dot redhat dot com
- Date: Fri, 10 Oct 2003 16:11:19 +0200
- Subject: RE: [ECOS] Probleme with ram rom data sections
> -----Message d'origine-----
> De: Gary Thomas [mailto:gary@mlbassoc.com]
> Date: vendredi 10 octobre 2003 15:42
> À: Matthieu.GIRARDIN@fr.thalesgroup.com
> Cc: ecos-discuss@sources.redhat.com
> Objet: RE: [ECOS] Probleme with ram rom data sections
>
>
> On Fri, 2003-10-10 at 07:20,
> Matthieu.GIRARDIN@fr.thalesgroup.com wrote:
> > > > Well good afternoon everybody !
> > > >
> > > > I tried to modify my memory layout. I want 4 Mo ROM at
> > > 0x4000000 and 2Mo RAM
> > > > at 0x000.
> > > > I modify my mlt files as follow :
> > > >
> > > > mlt_powerpc_mbx_rom.ldi:
> > > >
> > > > MEMORY
> > > > {
> > > > ram : ORIGIN = 0, LENGTH = 0x200000
> > > > rom : ORIGIN = 0x4000000, LENGTH = 0x400000
> > > > }
> > > >
> > > > SECTIONS
> > > > {
> > > > SECTIONS_BEGIN
> > > > SECTION_vectors (rom, 0x4000000, LMA_EQ_VMA)
> > > > SECTION_text (rom, ALIGN (0x4), LMA_EQ_VMA)
> > > > SECTION_fini (rom, ALIGN (0x4), LMA_EQ_VMA)
> > > > SECTION_rodata1 (rom, ALIGN (0x8), LMA_EQ_VMA)
> > > > SECTION_rodata (rom, ALIGN (0x8), LMA_EQ_VMA)
> > > > SECTION_fixup (rom, ALIGN (0x4), LMA_EQ_VMA)
> > > > SECTION_gcc_except_table (rom, ALIGN (0x1), LMA_EQ_VMA)
> > > > CYG_LABEL_DEFN(__reserved_vectors) = 0; . =
> > > > CYG_LABEL_DEFN(__reserved_vectors) + 0x3000;
> > > > CYG_LABEL_DEFN(__reserved_vsr_table) = ALIGN (0x1); . =
> > > > CYG_LABEL_DEFN(__reserved_vsr_table) + 0x200;
> > > > CYG_LABEL_DEFN(__reserved_virtual_table) = ALIGN (0x1); . =
> > > > CYG_LABEL_DEFN(__reserved_virtual_table) + 0x100;
> > > > SECTION_data (ram, ALIGN (0x10),
> FOLLOWING(.gcc_except_table))
> > > > //SECTION_data (ram, ALIGN (0x10), AT(0x4200000))
> > > > SECTION_sbss (ram, ALIGN (0x4), LMA_EQ_VMA)
> > > > SECTION_bss (ram, ALIGN (0x10), LMA_EQ_VMA)
> > > > CYG_LABEL_DEFN(__heap1) = ALIGN (0x8);
> > > > SECTIONS_END
> > > > }
> > > >
> > > >
> > > > mlt_powerpc_mbx_rom.h:
> > > >
> > > > #ifndef __ASSEMBLER__
> > > > #include <cyg/infra/cyg_type.h>
> > > > #include <stddef.h>
> > > >
> > > > #endif
> > > > #define CYGMEM_REGION_ram (0)
> > > > #define CYGMEM_REGION_ram_SIZE (0x200000)
> > > > #define CYGMEM_REGION_ram_ATTR (CYGMEM_REGION_ATTR_R |
> > > CYGMEM_REGION_ATTR_W)
> > > > #define CYGMEM_REGION_rom (0x4000000)
> > > > #define CYGMEM_REGION_rom_SIZE (0x400000)
> > > > #define CYGMEM_REGION_rom_ATTR (CYGMEM_REGION_ATTR_R)
> > > > #ifndef __ASSEMBLER__
> > > > extern char CYG_LABEL_NAME (__reserved_vectors) [];
> > > > #endif
> > > > #define CYGMEM_SECTION_reserved_vectors (CYG_LABEL_NAME
> > > > (__reserved_vectors))
> > > > #define CYGMEM_SECTION_reserved_vectors_SIZE (0x3000)
> > > > #ifndef __ASSEMBLER__
> > > > extern char CYG_LABEL_NAME (__reserved_vsr_table) [];
> > > > #endif
> > > > #define CYGMEM_SECTION_reserved_vsr_table (CYG_LABEL_NAME
> > > > (__reserved_vsr_table))
> > > > #define CYGMEM_SECTION_reserved_vsr_table_SIZE (0x200)
> > > > #ifndef __ASSEMBLER__
> > > > extern char CYG_LABEL_NAME (__reserved_virtual_table) [];
> > > > #endif
> > > > #define CYGMEM_SECTION_reserved_virtual_table (CYG_LABEL_NAME
> > > > (__reserved_virtual_table))
> > > > #define CYGMEM_SECTION_reserved_virtual_table_SIZE (0x100)
> > > > #ifndef __ASSEMBLER__
> > > > extern char CYG_LABEL_NAME (__heap1) [];
> > > > #endif
> > > > #define CYGMEM_SECTION_heap1 (CYG_LABEL_NAME (__heap1))
> > > > #define CYGMEM_SECTION_heap1_SIZE (0x400000 - (size_t)
> > > CYG_LABEL_NAME
> > > > (__heap1))
> > > >
> > > > Well now, my problem is :
> > > > when I read the nm output from my file generated
> > > (powerpc-eabi-nm) I can see
> > > > (result 1 just under) the right symbol __rom_data_start at
> > > the end of the
> > > > rom but when I transform it into S-Record file (Motorola
> > > standard file)
> > > > before to flash it I can see (result 2 just under) that
> > > there is nothing
> > > > after this adress; as if the data section was empty...
> > > > If someone can help it would be great.
> > > >
> > > > Result 1:
> > > >
> > > > 04023498 V typeinfo name for std::bad_cast
> > > > 040234a4 V typeinfo name for std::type_info
> > > > 040234b4 W raise
> > > > 04023570 A _etext
> > > > 0402442c r hex.2
> > > > 040258ac A __FIXUP_START__
> > > > 04025a1c A __EXCEPT_END__
> > > > 04025a1c A __EXCEPT_START__
> > > > 04025a1c A __FIXUP_END__
> > > > 04025a20 A __rom_data_start
> > > >
> > > > Result 2:
> > > >
> > > > S3190402598004023364040233680402336C040233700402337CC6
> > > > S3190402599404023380040233840402338804023394040233981E
> > > > S319040259A80402339C040233A0040233AC040233B0040233B476
> > > > S319040259BC040233B8040233BC040233C0040233C4040233C8EE
> > > > S319040259D0040233CC040233D0040233D4040233D8040233DC76
> > > > S319040259E4040233E0040233E4040233E8040233EC040233F0FE
> > > > S319040259F8040233F4040233F8040233FC040234000402340484
> > > > S30D04025A0C040234080000000044
> > > <-- the end
> > > > of the rom
> > > >
> > > > nothing in data section
> > > > S31900003300000000003B9ACA00000000640400AA040400AC044A
> > > <-- the
> > > > beginning of the ram
> > > > S319000033140400D0E40400D0EC0400AD900400AD980400B05C8D
> > > > S319000033280400A9380400B08C0400B1A00400B2600400B398AC
> > > > S3190000333C0400D0E40400D0EC0400B5440400B56C0400B5D450
> > >
> > > What does "powerpc-eabi-objdump -h <FILE>.elf" show?
> > > What does your command to create the S-record file look like?
> > >
> > > --
> > > Gary Thomas <gary@mlbassoc.com>
> > > MLB Associates
> > >
> >
> > Well, just under is the result of "powerpc-eabi-objdump -h
> <FILE>.elf".
> > My command to create my S-Record is "dump -Rv -o map.dmp
> <FILE>.elf".
> > So it seems the objdump give us a correct answer (or I'am
> stupid...).
> > But I can't load my elf in flash because before when I read
> my map.dmp
> > there is nothing in the data section.
> > Thanks if you can help.
> >
> > powerpc-eabi-objdump:
> >
> > perf.otr.rom.le.1010.a.1502.elf: file format elf32-powerpc
> >
> > Sections:
> > Idx Name Size VMA LMA File off Algn
> > 0 .vectors 00002000 04000000 04000000 00010000 2**8
> > CONTENTS, ALLOC, LOAD, READONLY, CODE
> > 1 .text 00021570 04002000 04002000 00012000 2**3
> > CONTENTS, ALLOC, LOAD, READONLY, CODE
> > 2 .fini 00000000 04023570 04023570 00045694 2**0
> > CONTENTS
> > 3 .rodata1 00000000 04023570 04023570 00045694 2**0
> > CONTENTS
> > 4 .rodata 0000233c 04023570 04023570 00033570 2**2
> > CONTENTS, ALLOC, LOAD, READONLY, DATA
> > 5 .fixup 00000170 040258ac 040258ac 000358ac 2**2
> > CONTENTS, ALLOC, LOAD, DATA
> > 6 .gcc_except_table 00000000 04025a1c 04025a1c 00045694 2**0
> > CONTENTS
> > 7 .data 00002394 00003300 04025a20 00043300 2**3
> > CONTENTS, ALLOC, LOAD, CODE
> > 8 .sbss 000000d5 00005694 00005694 00005694 2**2
> > ALLOC
> > 9 .bss 000f792c 00005770 00005770 0000569b 2**4
> > ALLOC
> > 10 .comment 0000061e 00000000 00000000 00045694 2**0
> > CONTENTS, READONLY
>
> Have you tried to use objcopy to create the S-record file?
> % powerpc-eabi-objcopy -O srec <FILE>.elf <FILE>.srec
> I know nothing of your "dump" program, but objcopy does the
> right thing.
>
> --
> Gary Thomas <gary@mlbassoc.com>
> MLB Associates
>
Ok it works, yeaaaaaaaaaaaaaaaaaaaaaaaah !
In fact I think that my gnu tool dump is too old to generate a good s-record
from your elf files. It works with your dedicaced tool "powerpc-eabi-objcopy
-O srec".
So a big thanks to you for your help.
(I believe you'are the autor of mbx files : gthomas ? no)
Big thanks for your help and for eCos !
--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss