[haiku-development] Re: Partition Table Woes (Was: Filesystem corruption and Alpha-1)
- From: Urias McCullough <umccullough@xxxxxxxxx>
- To: haiku-development@xxxxxxxxxxxxx
- Date: Wed, 1 Jul 2009 16:06:34 -0700
2009/7/1 André Braga <meianoite@xxxxxxxxx>:
> Em 01/07/2009, às 19:38, Urias McCullough <umccullough@xxxxxxxxx> escreveu:
>>
>> This doesn't mean the data is gone, and as long as the bootloader
>> still is there, and knows the offset of each partition, they should
>> still boot.
>
> I think I remember BeOS working like this, but Haiku won't go past
> haiku_loader unless the loader manages to parse at least one valid
> partitioning scheme and detect a BFS superblock. I know that first hand
> because of my fiddlings with the GPT partitioning scheme.
>
> Which is very funny, since, well, the loader is executed from a BFS volume.
> :)
Ah, you may be right - as Haiku does try to interpret the partition
map... But Haiku can also been booted from a partition-less drive -
E.G. a raw image dd'd over an entire disk - so there must be some
other magic in there as well to deal with situations where no
partition map exists?
In any case, if gparted can't find the partition map, things have gone
horribly wrong ;)
- Urias
Other related posts:
