* [ECOS] redboot startup mode @ 2005-05-24 18:26 yrodguez 2005-05-24 22:27 ` Gary Thomas 0 siblings, 1 reply; 6+ messages in thread From: yrodguez @ 2005-05-24 18:26 UTC (permalink / raw) To: ecos-discuss hi, i have a doubt/question? i'm trying to compile a redboot in romram mode. At http://sources.redhat.com/ecos/docs-latest/redboot/excaliburarm9.html say that romram mode runs redboot from ram with redboot in flash. I supossed that redboot is stored in flash memory and then it is copied from flash to ram. But when i desassemble the file i get the code is mapped on ram (0x8000): redboot.elf: file format elf32-littlearm Disassembly of section .rom_vectors: 00008000 <__exception_handlers>: 8000: ea00000e b 8040 <_stext> 8004: e59ff018 ldr pc, [pc, #24] ; 8024 <.undefined_instruction> 8008: e59ff018 ldr pc, [pc, #24] ; 8028 <.software_interrupt> 800c: e59ff018 ldr pc, [pc, #24] ; 802c <.abort_prefetch> 8010: e59ff018 ldr pc, [pc, #24] ; 8030 <.abort_data> 8014: 00000000 andeq r0, r0, r0 8018: e59ff018 ldr pc, [pc, #24] ; 8038 <.IRQ> 801c: e59ff018 ldr pc, [pc, #24] ; 803c <.FIQ> 00008020 <vectors>: 8020: 00000040 andeq r0, r0, r0, asr #32 00008024 <.undefined_instruction>: 8024: 00008494 muleq r0, r4, r4 00008028 <.software_interrupt>: 8028: 000084b8 streqh r8, [r0], -r8 etc.... Maybe i don`t understand this mode at all. thanks. -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [ECOS] redboot startup mode 2005-05-24 18:26 [ECOS] redboot startup mode yrodguez @ 2005-05-24 22:27 ` Gary Thomas [not found] ` <20050524175806.gj72jy7dy8kw0ck4@webmail.iuma.ulpgc.es> 0 siblings, 1 reply; 6+ messages in thread From: Gary Thomas @ 2005-05-24 22:27 UTC (permalink / raw) To: yrodguez; +Cc: eCos Discussion On Tue, 2005-05-24 at 17:41 +0100, yrodguez@iuma.ulpgc.es wrote: > hi, > i have a doubt/question? > > i'm trying to compile a redboot in romram mode. > > > At http://sources.redhat.com/ecos/docs-latest/redboot/excaliburarm9.html > > say that romram mode runs redboot from ram with redboot in flash. I supossed > that redboot is stored in flash memory and then it is copied from flash to ram. > But when i desassemble the file i get the code is mapped on ram (0x8000): > Correct. It has to be that way - the linker produces an image suitable for running from RAM. That said, the image will be stored in FLASH (typically this is all that's accessible immediately after RESET). The program [RedBoot or whatever] then does some minimal hardware initialization, e.g. turning on the SDRAM controller, setting up mapping registers, etc. The code is then copied from FLASH to its ultimate RAM destination and control is then passed there. Now you are running from RAM only. It's up to the startup code to understand all of this and it must contain [normally position-independent] code that performs these steps at startup. Note: this mode is only suitable for *the* initial "ROM" monitor code, like RedBoot. This is not how you would write a generic program which runs in RAM but whose image was stored in FLASH. For that, you use RedBoot's FIS commands (create, load) -- ------------------------------------------------------------ Gary Thomas | Consulting for the MLB Associates | Embedded world ------------------------------------------------------------ -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss ^ permalink raw reply [flat|nested] 6+ messages in thread
[parent not found: <20050524175806.gj72jy7dy8kw0ck4@webmail.iuma.ulpgc.es>]
* Re: [ECOS] redboot startup mode [not found] ` <20050524175806.gj72jy7dy8kw0ck4@webmail.iuma.ulpgc.es> @ 2005-05-25 2:37 ` Gary Thomas 2005-05-25 6:57 ` yrodguez 0 siblings, 1 reply; 6+ messages in thread From: Gary Thomas @ 2005-05-25 2:37 UTC (permalink / raw) To: yrodguez; +Cc: eCos Discussion Please reply to the eCos mailing list so that all may benefit. Private support is available under contract (from MLB Associates or others) On Tue, 2005-05-24 at 17:58 +0100, yrodguez@iuma.ulpgc.es wrote: > But, who exactly have to copy redboot code from flash to ram?, I'm Look in the platform startup code: <ecos>/packages/arm/arm9/excalibur/current/include/hal_platform_setup.h > sorry, maybe > i don't understand it due to my bad level in english > > thanks. > > Quoting Gary Thomas <gary@mlbassoc.com>: > > > On Tue, 2005-05-24 at 17:41 +0100, yrodguez@iuma.ulpgc.es wrote: > >> hi, > >> i have a doubt/question? > >> > >> i'm trying to compile a redboot in romram mode. > >> > >> > >> At http://sources.redhat.com/ecos/docs-latest/redboot/excaliburarm9.html > >> > >> say that romram mode runs redboot from ram with redboot in flash. I supossed > >> that redboot is stored in flash memory and then it is copied from > >> flash to ram. > >> But when i desassemble the file i get the code is mapped on ram (0x8000): > >> > > > > Correct. It has to be that way - the linker produces an image suitable > > for running from RAM. That said, the image will be stored in FLASH > > (typically this is all that's accessible immediately after RESET). The > > program [RedBoot or whatever] then does some minimal hardware > > initialization, e.g. turning on the SDRAM controller, setting up mapping > > registers, etc. The code is then copied from FLASH to its ultimate RAM > > destination and control is then passed there. Now you are running from > > RAM only. It's up to the startup code to understand all of this and > > it must contain [normally position-independent] code that performs these > > steps at startup. > > > > Note: this mode is only suitable for *the* initial "ROM" monitor code, > > like RedBoot. This is not how you would write a generic program which > > runs in RAM but whose image was stored in FLASH. For that, you use > > RedBoot's FIS commands (create, load) > > > > -- > > ------------------------------------------------------------ > > Gary Thomas | Consulting for the > > MLB Associates | Embedded world > > ------------------------------------------------------------ > > > > > > -- ------------------------------------------------------------ Gary Thomas | Consulting for the MLB Associates | Embedded world ------------------------------------------------------------ -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [ECOS] redboot startup mode 2005-05-25 2:37 ` Gary Thomas @ 2005-05-25 6:57 ` yrodguez 2005-05-25 7:50 ` [ECOS] Clean room module loader + GPL module + Application C.J. 0 siblings, 1 reply; 6+ messages in thread From: yrodguez @ 2005-05-25 6:57 UTC (permalink / raw) To: ecos-discuss Quoting Gary Thomas <gary@mlbassoc.com>: > Please reply to the eCos mailing list so that all may benefit. Private > support is available under contract (from MLB Associates or others) I'm Sorry. > > On Tue, 2005-05-24 at 17:58 +0100, yrodguez@iuma.ulpgc.es wrote: >> But, who exactly have to copy redboot code from flash to ram?, I'm > > Look in the platform startup code: > <ecos>/packages/arm/arm9/excalibur/current/include/hal_platform_setup.h > My question is: do i need another startup code, different to redboot, to use this mode or is enough with redboot? thanks, and, I'm sorry again. >> sorry, maybe >> i don't understand it due to my bad level in english >> >> thanks. >> >> Quoting Gary Thomas <gary@mlbassoc.com>: >> >> > On Tue, 2005-05-24 at 17:41 +0100, yrodguez@iuma.ulpgc.es wrote: >> >> hi, >> >> i have a doubt/question? >> >> >> >> i'm trying to compile a redboot in romram mode. >> >> >> >> >> >> At http://sources.redhat.com/ecos/docs-latest/redboot/excaliburarm9.html >> >> >> >> say that romram mode runs redboot from ram with redboot in flash. >> I supossed >> >> that redboot is stored in flash memory and then it is copied from >> >> flash to ram. >> >> But when i desassemble the file i get the code is mapped on ram (0x8000): >> >> >> > >> > Correct. It has to be that way - the linker produces an image suitable >> > for running from RAM. That said, the image will be stored in FLASH >> > (typically this is all that's accessible immediately after RESET). The >> > program [RedBoot or whatever] then does some minimal hardware >> > initialization, e.g. turning on the SDRAM controller, setting up mapping >> > registers, etc. The code is then copied from FLASH to its ultimate RAM >> > destination and control is then passed there. Now you are running from >> > RAM only. It's up to the startup code to understand all of this and >> > it must contain [normally position-independent] code that performs these >> > steps at startup. >> > >> > Note: this mode is only suitable for *the* initial "ROM" monitor code, >> > like RedBoot. This is not how you would write a generic program which >> > runs in RAM but whose image was stored in FLASH. For that, you use >> > RedBoot's FIS commands (create, load) >> > >> > -- >> > ------------------------------------------------------------ >> > Gary Thomas | Consulting for the >> > MLB Associates | Embedded world >> > ------------------------------------------------------------ >> > >> > >> >> > -- > ------------------------------------------------------------ > Gary Thomas | Consulting for the > MLB Associates | Embedded world > ------------------------------------------------------------ > > -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss ^ permalink raw reply [flat|nested] 6+ messages in thread
* [ECOS] Clean room module loader + GPL module + Application 2005-05-25 6:57 ` yrodguez @ 2005-05-25 7:50 ` C.J. 2005-05-26 12:30 ` Bart Veer 0 siblings, 1 reply; 6+ messages in thread From: C.J. @ 2005-05-25 7:50 UTC (permalink / raw) To: ecos-discuss Hi, I know eCos mod-GPL have been discussed several times. I also understand that eCos is not GPL compatible --- Using GPL code in eCos kernel will force the license of eCos application GPL. GNU GPL (http://www.gnu.org/copyleft/gpl.html) states: The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. Considering only legal issues (just ignore any technical or code size issue), If an eCos system is implemented like this: (a) eCos kerenl (mod-GPL) (b) Clean room module loader (closed license) (c) A GPL module registers itself as "/dev/storage" (GPL) (d) Clean room application, mount "/dev/storage" and use eCos POSIX layer (via FAT filesystem) to access the device (closed license). Does the GPL "special exception" apply to this situation? Is it legal If the provider do not release the source code of (b) and (d)? Best Regards, C.J. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [ECOS] Clean room module loader + GPL module + Application 2005-05-25 7:50 ` [ECOS] Clean room module loader + GPL module + Application C.J. @ 2005-05-26 12:30 ` Bart Veer 0 siblings, 0 replies; 6+ messages in thread From: Bart Veer @ 2005-05-26 12:30 UTC (permalink / raw) To: tsai.cj; +Cc: ecos-discuss >>>>> " " == Retallack, Mark (Siemens) <mark.retallack@siemens.com> writes: > I know eCos mod-GPL have been discussed several times. I also > understand that eCos is not GPL compatible Incorrect. The old RHEPL license was incompatible with the GPL. The current GPL+exception license is compatible. > --- Using GPL code in eCos kernel will force the license of > eCos application GPL. The reference to the eCos kernel is irrelevant. Using GPL code means that you have to comply with the terms of that license. If your application makes use of two other bits of code released under different licenses then it must comply with the terms of both licenses. You cannot ignore the license of one just because the other happens to be released on more liberal terms. For an analogy, suppose your application makes use of a proprietary OS and a proprietary device driver supplied by a different software company. You would expect to pay two license fees, one to the OS company and one to the company providing the driver. You cannot avoid paying the OS license fee because you are paying somebody else for different software. <snip> > Considering only legal issues (just ignore any technical or > code size issue), You cannot get a definitive answer to legal questions on this mailing list. Instead you need to consult a copyright lawyer who understands these issues, including legal variations between all the countries where you expect to do business. > If an eCos system is implemented like this: > (a) eCos kerenl (mod-GPL) > (b) Clean room module loader (closed license) > (c) A GPL module registers itself as "/dev/storage" (GPL) > (d) Clean room application, mount "/dev/storage" and use eCos > POSIX layer (via FAT filesystem) to access the device (closed > license). > Does the GPL "special exception" apply to this situation? Is it > legal If the provider do not release the source code of (b) and > (d)? My understanding is that (b) is fine. Such a module loader would be equivalent to application code and hence can be kept proprietary. However (d) is not fine. The application depends on the GPL'd code. My understanding is that legally this makes it a derived work and hence subject to the terms of the GPL. You cannot just bypass the GPL by adding a few indirections, copyright law does not work like that. Similarly under Linux you cannot just take GPL'd code like the gcc code generator, turn it into a shared library, and then link your proprietary application with the shared library. And before people start quoting the Linux kernel as a counter example, it has an explicit exemption allowing application code to use kernel services by the normal Linux system calls. See the top level COPYING file in the kernel sources. No such exemption applies in your scenario. I am not a lawyer and this is not official legal advice. If in doubt then you need to consult somebody suitably qualified to answer these questions. Bart -- Bart Veer eCos Configuration Architect http://www.ecoscentric.com/ The eCos and RedBoot experts -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2005-05-25 14:30 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2005-05-24 18:26 [ECOS] redboot startup mode yrodguez 2005-05-24 22:27 ` Gary Thomas [not found] ` <20050524175806.gj72jy7dy8kw0ck4@webmail.iuma.ulpgc.es> 2005-05-25 2:37 ` Gary Thomas 2005-05-25 6:57 ` yrodguez 2005-05-25 7:50 ` [ECOS] Clean room module loader + GPL module + Application C.J. 2005-05-26 12:30 ` Bart Veer
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).