[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ok i threw the main ideas, now i must find the time to code it correctly. Cry and shout loud on this ML if i did nothing until the end of the week. Sounds good to me. Don't think that was a dig, I totally understand if you don't have so much time -- I experience that too :-). Out of interest, do Thomson notify you of the changes, or do they rely on the community to reverse engineer it? I get the impression that for a product which sports a Supported in GNU/Linux they don't do much to help it. I must say I dream of a day when we have completely non-proprietary firmware... Regards, Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAKAut3bF34WXq1XsRAmS0AKCnr3CapF00Bkru8ZmTf4zU9lv2zwCffxRa bbMBzBHizgwXGTyuezRzk3A= =U5PR -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Hello Martin, still appears mangled. Sorry :( . It's fixed now! Sort of. I think the original source doesn't use all the spaces but tabs instead. Could you be so kind to upload a cleanly patching diff instead of a hand crafted version? diff -u modem_run.c.000 modem_run.c I think in general it is good for maintenance if indention is preserved carefully. Tell your editor to use 8 space tabs instead of 8 spaces. If you are using the midnight commander disable the fake half tabs. Bye, Leonard. -- mount -t life -o ro /dev/dna /genetic/research Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Hello Martin, Sort of. I think the original source doesn't use all the spaces but tabs instead. Could you be so kind to upload a cleanly patching diff instead of a hand crafted version? diff -u modem_run.c.000 modem_run.c I think in general it is good for maintenance if indention is preserved carefully. Tell your editor to use 8 space tabs instead of 8 spaces. If you are using the midnight commander disable the fake half tabs. Got the tarball. The problem appears to be some spaces at the end of some lines that are not taken into account in your hand patch. (Maybe only in my version of the source and already fixed in CVS. If the original diff patches cleanly for you please let me know.) Please do a diff the next time, as it avoids these kind of nasty problems. Also the second block uses spaces instead of tabs, 8 or 9 spaces alternatingly. I am fixing that as well. A clean patch will be available within 24 hours from http://www.ottolander.nl/opensource/speedtouch/modem_run.c.diff . Bye, Leonard. -- mount -t life -o ro /dev/dna /genetic/research Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tuesday 10 February 2004 11:08 pm, Leonard den Ottolander wrote: Hello Martin, Sort of. I think the original source doesn't use all the spaces but tabs instead. Could you be so kind to upload a cleanly patching diff instead of a hand crafted version? diff -u modem_run.c.000 modem_run.c I think in general it is good for maintenance if indention is preserved carefully. Tell your editor to use 8 space tabs instead of 8 spaces. If you are using the midnight commander disable the fake half tabs. Got the tarball. The problem appears to be some spaces at the end of some lines that are not taken into account in your hand patch. (Maybe only in my version of the source and already fixed in CVS. If the original diff patches cleanly for you please let me know.) Please do a diff the next time, as it avoids these kind of nasty problems. Also the second block uses spaces instead of tabs, 8 or 9 spaces alternatingly. I am fixing that as well. A clean patch will be available within 24 hours from http://www.ottolander.nl/opensource/speedtouch/modem_run.c.diff . Bye, Leonard. I've uploaded a fresh diff now. I apoligise for this, the patch I had uploaded was what I had been sent, I didn't recreate it. Hopefully it's alright now. Regards, Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAKW9L3bF34WXq1XsRAv1QAKD2XVvYOSDfZO0/0mtNaMULCt6cfQCgmFRn 2j+aEZECZUjIGwhJu8QVN6A= =1Wt2 -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wednesday 11 February 2004 12:18 am, Leonard den Ottolander wrote: Hi Martin, Also the second block uses spaces instead of tabs, 8 or 9 spaces alternatingly. I am fixing that as well. A clean patch will be available within 24 hours from http://www.ottolander.nl/opensource/speedtouch/modem_run.c.diff . My version is available. Created by diffing the tarball against the old tarball and cleaning up the spacing. I've uploaded a fresh diff now. I apoligise for this, the patch I had uploaded was what I had been sent, I didn't recreate it. Sent by whom? Who is the creator of that patch? Does that patch patch cleanly for you (patch -p 0 modem_run.diff)? Maybe the offending spaces have been deleted from CVS, in which case I am mistaken. The patch is perfectly clean for me. Either way, to finish this now, I'll upload your patch. I am the author of the patch, though what was originally uploaded was sent to me by Remigio Di Muzio (without him realising I was the author). http://www.nemohackers.org/speedtouch/modem_run.diff is now your diff. Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAKXcR3bF34WXq1XsRAtzmAKDc7MwxbNb2vCmoSyFzQ3MbJIQUMQCgyWJW /ppB2CJ0aHxrM7pkRnmlBBw= =9jKF -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Hello Martin, The patch is perfectly clean for me. Either way, to finish this now, I'll upload your patch. I am the author of the patch, though what was originally uploaded was sent to me by Remigio Di Muzio (without him realising I was the author). g http://www.nemohackers.org/speedtouch/modem_run.diff is now your diff. Now you are moving to fast ;) . If the patch you got patches cleanly against CVS mine won't. The first hunk fixes two redundant spaces at the end of the lines. It is needed for the tarball that I distribute inside my rpm, but not necessarily valid against current CVS. So if you have problems using this version the first hunk can safely be dropped. The spacing of the last block is nicer however. Bye, Leonard. -- mount -t life -o ro /dev/dna /genetic/research Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Hello Martin, I finally got my own patch to work today! So, I'm free once again. http://www.nemohackers.org/speedtouch.php If I apply this patch to modem_run will it still work with the original 330? If so did you add it to CVS? If not, how do you expect to merge it? An extra command line switch? Bye, Leonard. -- mount -t life -o ro /dev/dna /genetic/research Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Monday 09 February 2004 09:59 pm, Leonard den Ottolander wrote: Hello Martin, I finally got my own patch to work today! So, I'm free once again. http://www.nemohackers.org/speedtouch.php If I apply this patch to modem_run will it still work with the original 330? If so did you add it to CVS? If not, how do you expect to merge it? An extra command line switch? Bye, Leonard. No, it wont work with any 330 before the Rev 0400, which is what I explicitly stated on the page, and one of the reasons why it _isn't_ committed to CVS. A command line switch would be possible, though the thread help needed with testing auto-detect Rev patch is working on automatic detection -- the better option. If this happens, the ARM bytes from the patch on the site above can be merged with theirs, and finally merged with CVS. Regards, Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAKAQz3bF34WXq1XsRAqagAKDs+CE/tc/GpXD0VKlTbAC2yao4HQCg6t40 43Q46dNX1z/PMBbIhOeLaUY= =Qfnh -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Martin Galpin ([EMAIL PROTECTED]) wrote:
Until a real fix is made, I wrote a short page for new users on how to
get the Rev 400 modem to work.
This mail is just posted to show i'm not ignoring this nice thread and
that i'm porting attention to it. It's just i have not so much time
left to code these days. So i'll explain my plan.
Ok i started hacking a new firmware extractor. Here i sumarize the
needed steps:
1 - Reverse the modem_run design to old 1.1 design, as this works with
both new versions and old versions.
2 - Write a new firmware extractor.
1 and 2 are pretty independent when it comes to old modems. New modems
would need 2 *and* 1.
Basically, the new firmware extractor will look into the firmware file
for two ARM codes, the bootstrap code and the proper firmware. This will
work well for old firmware files that shipped both codes. With newer
releases, thomson separated the codes in two files, so i'll extend the
command line so you can tell the firmware loader to load the two pieces
of code from two different files.
I'll also add a bootstrap code CRC checking to detect if we try to
bootstrap a rev 4 with a rev 4 bootstrap code, and vice versa. This
will be a very similar code to the one that checks the big firmware.
It will all use a struct like:
typedef struct
{
unsigned char *phase1; /* Boostrap code */
unsigned char *phase2; /* Firmware */
unsigned long phase1_length;
unsigned long phase2_length;
} stusb_firmware_image_t;
The function i'm coding will extend the current firmware loader and
will look like:
stusb_firmware_image_t *firmware_load(char *phase1_filename,
char *phase2_filename);
The function behavior will be as follow:
- if phase1_filename == NULL then phase1_filename = phase2_filename;
(old firmware used to contain both codes)
- allocate a stusb_firmware_image_t *img;
- search boostrap code into phase1_filename, if not found
img-phase1 = NULL, if copy the data block and set its size.
- same for phase2
- return img
Then in modem_run, if img-phase1 is null, we set it to the free
bootstrap code that we can distribute safely in our code.
Then, step 3, check the modem revision, and identify the boostrap code
to determine which version it can bootstrap. If the modem revision
doesn't match the boostrap code, shout loud and exit, else continue.
I think that's all i have in mind atm. Ah Thomson is probably going to
change the firmware files formats in next releases, so this design
should help in supporting this new type of files in the future (unless
thomson decides to back the file format change).
Ok i threw the main ideas, now i must find the time to code it
correctly. Cry and shout loud on this ML if i did nothing until the end
of the week.
--
Edouard Gomez
Liste de diffusion modem ALCATEL SpeedTouch USB
Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ok i threw the main ideas, now i must find the time to code it correctly. Cry and shout loud on this ML if i did nothing until the end of the week. Sounds good to me. Don't think that was a dig, I totally understand if you don't have so much time -- I experience that too :-). Out of interest, do Thomson notify you of the changes, or do they rely on the community to reverse engineer it? I get the impression that for a product which sports a Supported in GNU/Linux they don't do much to help it. I must say I dream of a day when we have completely non-proprietary firmware... Regards, Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAKAwQ3bF34WXq1XsRAoq4AKC87sr4GmhLG7lAcnKqcmBGDNx0TwCfcKX2 sievJfgLkYYAZH21mkP54Vk= =HoYD -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Martin Galpin ([EMAIL PROTECTED]) wrote: Out of interest, do Thomson notify you of the changes, or do they rely on the community to reverse engineer it? I get the impression that for a product which sports a Supported in GNU/Linux they don't do much to help it. Since the first 330 is out, a thomson person contacted me with a patch for our modem_run program and pppoa3. And for this last speedtouch revision, benoit, francois, duncan and I have received sample modems from thomson to test our changes. If all goes well, we'll even have some striped down specs so we finally understand all the voodoo behind the usb control/interface usb code. Last year thomson did fail in providing us these docs, but i still hope, thomson will finally provide us these important docs. I don't have a ready version yet, because the last patches they provided us are just incompatible with old revisions, and their firmware extractor is only adpated to the new file format they'll probably adopt. At this moment, they asked me not to publish it, as they didn't make their choice yet on the format adoption (and it would not be so useful to end users if the fileformat isn't used ;-) I'd say, they support us enough so the modems work. It's not the ideal solution, but they support GNU/linux (and indirectly FreeBSD) more than some others USB ADSL modem manufacturers. -- Edouard Gomez Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sunday 08 February 2004 03:37 pm, Leonard den Ottolander wrote: Hello Martin, I finally got my own patch to work today! So, I'm free once again. Great! Thanks for putting in your effort. http://www.nemohackers.org/speedtouch.php The patch available there seems to be mangled. Bye, Leonard. So it does. I apoligise for that, I think gFTP messed up alittle. It's fixed now. Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAJltA3bF34WXq1XsRAin5AKC8wQWmoK1mMHdRwjRtZU2roYvIGACgyg1j XmVj9z4Ds7yD8sv1qpbg94w= =HUMb -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
Hello Martin, http://www.nemohackers.org/speedtouch.php The patch available there seems to be mangled. So it does. I apoligise for that, I think gFTP messed up alittle. It's fixed now. http://www.nemohackers.org/speedtouch/ : modem_run.diff 08-Feb-2004 10:4219k still appears mangled. Sorry :( . Bye, Leonard. -- mount -t life -o ro /dev/dna /genetic/research Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
[speedtouch] Re: 330/Rev 400 success
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sunday 08 February 2004 04:24 pm, Leonard den Ottolander wrote: Hello Martin, http://www.nemohackers.org/speedtouch.php The patch available there seems to be mangled. So it does. I apoligise for that, I think gFTP messed up alittle. It's fixed now. http://www.nemohackers.org/speedtouch/ : modem_run.diff 08-Feb-2004 10:4219k still appears mangled. Sorry :( . It's fixed now! Sorry for that. Martin - -- Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFAJmhV3bF34WXq1XsRAuM8AKD1n+OJPlDTIiS96KBlcWEb39p77gCfV0Bv JYy/f7zFUCmGt2iBDv7L0KE= =91Rq -END PGP SIGNATURE- Liste de diffusion modem ALCATEL SpeedTouch USB Pour se désinscrire : mailto:[EMAIL PROTECTED]
