bug-xorriso
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-xorriso] Test Proposal for [bug #46716] Protective MBR partitio


From: Thomas Schmitt
Subject: Re: [Bug-xorriso] Test Proposal for [bug #46716] Protective MBR partition is not marked as bootable
Date: Sun, 27 Dec 2015 14:06:42 +0100

Hi,

i wrote:
> > Awaiting your results and Andrei's opinion.
> > [about MBR dummy partition with boot flag]

Andrei Borzenkov:
> TBH I think that the best would be generic support for partition table
> manipulation 

The more generic it is, the more complicated it is to coordinate
with the various use case specific automats. 

I am slowly working towards the building stones for implementing
a language that describes System Area (the first 16 blocks of the ISO)
and El Torito.
It must enable the user to express all appropriate wishes,
and it must be suitable for the command line.

I will ponder whether i can already define the subset of that language
which is needed to implement the use case of synthetic partition table.

There are already two more or less failed attempts of such a language:
- The mkisofs option interface as used by grub-mkrescue and most
  other boot image producers.
- The xorriso command -boot_image which can express everything
  that the mkisofs options can. Just more verbous. The intended
  structure of specialization on particular boot loaders did not
  pay off. The moving target turned out to be not new boot loaders
  but old and new boot firmwares.

Both suffer from being designed mainly for El Torito and opaque
blobs for the System Area.
Weaving a net of boot sectors and partition tables has added
spaghetti noodle by spaghetti noodle.

There is a classic monument for all programmers who fell victim
to such a design hairball:
  
https://de.wikipedia.org/wiki/Datei:Laocoon_Pio-Clementino_Inv1059-1064-1067.jpg

The movements of boot firmware seems to slow down. It is about
time to re-model. Language design is not really my thing, though.


> It could be modeled after boot support,

"Boot support". Is that a particular technical term ?
https://www.gnu.org/software/grub/manual/html_node/Filesystem.html#Filesystem
?

I could compile a substantial list of contraints and examples
for the model which shall cover all boot firmware aspects of
an ISO.
If anybody is interested ... just ask.


Have a nice day :)

Thomas




reply via email to

[Prev in Thread] Current Thread [Next in Thread]