Hi, Miller, Shao:> I am enjoying this discussionMe not. I feel quite misunderstood. hpa:> Actually, I would prefer if you removed this in its current form.Ok. Will do. If any viable solution emerges, please let me know. Have a nice day :) Thomas
Uhh... Thomas, I think you missed the double-meaning at the bottom of Syslinux mailing-list e-mails... "Please do not send private replies to mailing list traffic." Here's the entire e-mail you quoted me from: ----- I am enjoying this discussion and trying to learn. Thank you. - Shao P. S. I do APM/GPT/MBR hybrid stuff by hand as well as MEMDISK's El Torito emulation, and am trying to understand what the challenge(s) are here. ----- Anyway, was the El Torito emulation idea something that might be straight-forward to rip-off from MEMDISK? - Shao Miller
Hi,> Uhh... Thomas, I think you missed the double-meaning at the bottom of > Syslinux mailing-list e-mails...Duh. The subject Subject: RE: [syslinux] port syslinux isohybrid perl script to C made me think it came from this list. Sorry for messing up.> Anyway, was the El Torito emulation idea something that might be > straight-forward to rip-off from MEMDISK?This is rather a question to hpa, i guess. My own expertise is limited to the production of ISO images. I can predict the LBA of file isolinux.bin and the image size. I could take a blob of at most 32 KB, perform some prescribed computations, insert the results at some prescribed byte addresses, and write the patched blob as the beginning of the emerging ISO image. Currently i lack of blob and prescriptions. There are no known problems with ElTorito and ISOLINUX. So the ElTorito record produced by libisofs should be able to help the blob with finding isolinux.bin, if desired. Have a nice day :) Thomas
Hello all,
Past few days I was working on porting the isohybrid perl script to c.
(-> https://fedorahosted.org/fedora-engineering-services/ticket/15)
The first patch towards the same is attached with this mail. The same could also
be accessed
from: http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch
I would really appreciate it if somebody could have a look at it and let me know
the feedback.
Thank you.
---
Regards
-P J P
PS: Please don't send me html/attachment/Fwd mails
Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download
Now! http://messenger.yahoo.com/download.php
-------------- next part --------------
A non-text attachment was scrubbed...
Name: syslinux-3.86-isohybrid.patch
Type: application/octet-stream
Size: 29071 bytes
Desc: not available
URL:
<http://www.zytor.com/pipermail/syslinux/attachments/20100411/4c933b0c/attachment.obj>
Hi, Please have a look at the new patch attached with this mail, and also accessible from: http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch.1 I've made most of the changes you suggested, - unit8_t, uint16_t, and u_int32_t. - The hexdata[] array is defined in hexdata.c created by make - bin2hex.pl now outputs data as C array. --- On Tue, 13/4/10, H. Peter Anvin <hpa at zytor.com> wrote:> Yes, I'm bloody sure! If you look at the documentation for > [s]printf you'll even find that %hh... is meaningless for printf > (as opposed to scanf).So, how does one write binary byte sequence, write(2)?> > About testing it, I tested it against > the F12 boot.iso, with both your Perl script as well as the > .c program, and both of them show the *same* values for > variables when printed. > > Clearly you didn't actually test it properly, since there > is simply no way the code could have output the proper result.Well ok, so how do I test it properly? Thank you. --- Regards -Prasad PS: Please don't send me html/attachment/Fwd mails Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download Now! http://messenger.yahoo.com/download.php -------------- next part -------------- A non-text attachment was scrubbed... Name: syslinux-3.86-isohybrid.patch.1 Type: application/x-troff-man Size: 13679 bytes Desc: not available URL: <http://www.zytor.com/pipermail/syslinux/attachments/20100414/02d8bbb1/attachment.1>
On 04/13/2010 11:32 AM, P J P wrote:> > So, how does one write binary byte sequence, write(2)? >fwrite(3) is the best option. Also keep in mind that you might be on a system of different native byte order.>>> About testing it, I tested it against >> the F12 boot.iso, with both your Perl script as well as the >> .c program, and both of them show the *same* values for >> variables when printed. >> >> Clearly you didn't actually test it properly, since there >> is simply no way the code could have output the proper result. > > Well ok, so how do I test it properly?Try using cmp(1) against a fresh mkisofs'd image versus isohybrid. -hpa
Hi,
Please have a look at the revised patch attached with this mail, and also
accessible
at: http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch.2
The major changes:
- sprintf replaced with memcpy(3)
- open/read/write/close replaced with fopen/fread/fwrite/fclose resp.
Thank you.
---
Regards
-P J P
PS: Please don't send me html/attachment/Fwd mails
-------------- next part --------------
A non-text attachment was scrubbed...
Name: syslinux-3.86-isohybrid.patch.2
Type: application/x-troff-man
Size: 15578 bytes
Desc: not available
URL:
<http://www.zytor.com/pipermail/syslinux/attachments/20100415/9cc304ed/attachment.2>
Hi,
Please have a look at the latest patch attached with this mail and also
accessible
from: http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch.3
Major changes:
- Fixes a small glitch in reading `id' from offset 440 in the MBR.
- Adds a new script bin2Carray.pl to create hexdata.c; ie. bin2hex.pl
is untouched.
- Adds a new function to display_mbr() when using verbose mode.
Also, I did a bit of testing with the original Perl script and the this patch,
and it looks quite the same.
(tmp @ 14:32)$ cmp boot.iso boot.iso.1
boot.iso boot.iso.1 differ: byte 3, line 1
(tmp @ 14:32)$ cmp boot.iso boot.iso.2
boot.iso boot.iso.2 differ: byte 3, line 1
(tmp @ 14:34)$ cmp output.iso output.iso.1
output.iso output.iso.1 differ: byte 433, line 4
(tmp @ 14:35)$ cmp output.iso output.iso.2
output.iso output.iso.2 differ: byte 433, line 4
Thank you.
---
Regards
-P J P
PS: Please don't send me html/attachment/Fwd mails
-------------- next part --------------
A non-text attachment was scrubbed...
Name: syslinux-3.86-isohybrid.patch.3
Type: application/x-troff-man
Size: 16883 bytes
Desc: not available
URL:
<http://www.zytor.com/pipermail/syslinux/attachments/20100415/9c8d68f8/attachment.3>
Hi, --- On Thu, 29/4/10, H. Peter Anvin <hpa at zytor.com> wrote:> You need to fix your endianness problems, at the very least.I tried to fix the endianness problem, please have a look at the latest patch accessible at: - http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch.4 - https://fedorahosted.org/fedora-engineering-services/ticket/15 It does work on both types of machines now, I tested it on a ppc machine. (~ @ 13:05)$ ./isohybrid.pl output.iso.1 (~ @ 13:05)$ ./isohybrid output.iso.2 (~ @ 13:05)$ ./endian big endian n: 04030201 Addr: 3220897608|3220897609|3220897610|3220897611| Byte: 4| 3| 2| 1| (~ @ 13:05)$ cmp output.iso output.iso.1 output.iso output.iso.1 differ: byte 433, line 4 (~ @ 13:06)$ cmp output.iso output.iso.2 output.iso output.iso.2 differ: byte 433, line 4 Thanks. --- Regards -P J P
???Hi, --- On Thu, 29/4/10, H. Peter Anvin <hpa at zytor.com> wrote:> You need to fix your endianness problems, at the very least.I tried to fix the endianness problem, please have a look at the latest patch accessible at: - http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch.4 - https://fedorahosted.org/fedora-engineering-services/ticket/15 It does work on both types of machines now, I tested it on a ppc machine. (~ @ 13:05)$ ./isohybrid.pl output.iso.1 (~ @ 13:05)$ ./isohybrid output.iso.2 (~ @ 13:05)$ ./endian big endian n: 04030201 Addr: 3220897608|3220897609|3220897610|3220897611| Byte:? ? ? ? ? 4|? ? ? ???3|? ? ? ???2|? ? ? ???1| (~ @ 13:05)$ cmp output.iso output.iso.1 output.iso output.iso.1 differ: byte 433, line 4 (~ @ 13:06)$ cmp output.iso output.iso.2 output.iso output.iso.2 differ: byte 433, line 4 Thanks. --- Regards ? ? -P J P
On 05/04/2010 01:57 AM, Geert Stappers wrote:> Op 20100504 om 12:10 schreef P J P: >> >> Please see: http://pjp.dgplug.org/tools/syslinux-3.86-isohybrid.patch.8 >> > > For "compatebility tests" is it good > to have the original Perl script availble: >Not to mention cross-platform utility. -hpa