On Mon, Apr 3, 2023 at 5:31 AM Thomas Schmitt <scdbac...@gmx.net> wrote:
> Hi, > > Rocky Bernstein wrote: > > if xoriso or some other burner program in the past created ISO 9660 > images > > in a particular way, > > xorriso still produces the larger ISOs by default. Normally it does not > matter to have 300+ KiB of waste, while the advantage is safety against > the Linux TAO CD bug and the capability for recording a multi-session > history. > > Whatever, do i get it right that you want me to add the old wasteful ISO > to test/check_multiextent.sh like this ? > > for i in multi_extent_8k.iso \ > multi_extent_8k_big.iso \ > multi_extent_8k_joliet.iso > do > ... test ... > > Yes, please do that. Thanks. > > Already finding a good name for the "_big" ISO is difficult. I am > It is not that difficult "_big" or if you want to be more correct "_larger" is fine. > reluctant to call it "_wasteful" because it's waste only in the context of > libcdio tests. "_old" would be wrong too, because it is still the default > with xorriso. But "_default" would open the question what's wrong with > it or with the other non-default ISO. > It boils down to the fact that i cannot give a convincing motivation for > its existence. > The motivation is that these are different. Even though xorriso doesn't create the smaller one some other burner might. > > A compromise would be to replace both by an ISO which just lacks the end > padding but has the multi-session preparation: > This feels like more complication than compromise. > > dd bs=2048 count=96 if=multi_extent_8k_big.iso \ > of=multi_extent_8k.iso > > This would remove the end padding which is not counted as part of the ISO > filesystem (other than results from the mkisofs emulation). The result has > 196608 bytes rather than 524288. > People might wonder why the ISO for the Joliet test is still smaller. > Add a README file inside the "test" that describes the images, what they are and how they were created. That would probably be a good thing to do anyway. > > Have a nice day :) > > Thomas > > >
