Added a bit of logic to possibly fall back to a fixed 512 block MBR in case
certain conditions indicate that we might deal with a fixed 512 bytes MBR on
a non 512 byte block size medium. One condition is that at least one partition
had to be shrunk to fit the available size (which usually happens with larger
block sizes). We retry with a fixed 512 block size once and compare the result
if it didn't improve the situation it is reverted again. This is mostly a
preparation for the upcoming "anyboot" hybrid MBR/ISO images.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35664 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Removed SECTOR_SIZE in the read-only part of the intel partition scanner;
however, it's still used in the write part.
* Made the read-only code block size agnostic. Only tested with an Apple
iPod so far, and it recognizes its partition fine now. Next test on real
hardware.
* Cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30155 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Renamed partition_table_sector to partition_table, and PTS* to
PartitionTable*; the sector is not really a sector, but only the first
512 bytes of it.
* Cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30154 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added the partition size as parameter to the file and partitioning
system initialize() hooks. It's often the only info about the
partition one needs and thus locking the partition just to get it is
no longer necessary.
* intel partitioning system:
- Removed passing around block sizes. We require 512 byte sectors
anyway. In fact using the parent partition's block size was even
wrong.
- Simplified writing the partition map sector.
- Simplified and corrected the partition map initialization.
- We don't fail identifying a partition anymore, if the partition map
contains no partitions. We would never identify a freshly
initialized partition map before.
- Made pm_identify() more intelligent: It determines the priority to
return depending on whether the partition is the device itself and
whether we have recognized child partitions.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22447 a95241bf-73f2-0310-859d-f6bbb57e9c96

Write support for the Intel partitioning system module. Courtesy of
Tomas Kucera.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21697 a95241bf-73f2-0310-859d-f6bbb57e9c96

* When run as part of the boot loader, it will now adjust the size of partitions
to fit into the session - this should help bug #238 to disappear.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18988 a95241bf-73f2-0310-859d-f6bbb57e9c96

Refactored a bit -- pulled the partition map parser out. Userland add-on disabled for the time being, due to build problems.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9158 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added a bit of logic to possibly fall back to a fixed 512 block MBR in case
certain conditions indicate that we might deal with a fixed 512 bytes MBR on
a non 512 byte block size medium. One condition is that at least one partition
had to be shrunk to fit the available size (which usually happens with larger
block sizes). We retry with a fixed 512 block size once and compare the result
if it didn't improve the situation it is reverted again. This is mostly a
preparation for the upcoming "anyboot" hybrid MBR/ISO images.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35664 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Removed SECTOR_SIZE in the read-only part of the intel partition scanner;
however, it's still used in the write part.
* Made the read-only code block size agnostic. Only tested with an Apple
iPod so far, and it recognizes its partition fine now. Next test on real
hardware.
* Cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30155 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Renamed partition_table_sector to partition_table, and PTS* to
PartitionTable*; the sector is not really a sector, but only the first
512 bytes of it.
* Cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30154 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added the partition size as parameter to the file and partitioning
system initialize() hooks. It's often the only info about the
partition one needs and thus locking the partition just to get it is
no longer necessary.
* intel partitioning system:
- Removed passing around block sizes. We require 512 byte sectors
anyway. In fact using the parent partition's block size was even
wrong.
- Simplified writing the partition map sector.
- Simplified and corrected the partition map initialization.
- We don't fail identifying a partition anymore, if the partition map
contains no partitions. We would never identify a freshly
initialized partition map before.
- Made pm_identify() more intelligent: It determines the priority to
return depending on whether the partition is the device itself and
whether we have recognized child partitions.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22447 a95241bf-73f2-0310-859d-f6bbb57e9c96

Write support for the Intel partitioning system module. Courtesy of
Tomas Kucera.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21697 a95241bf-73f2-0310-859d-f6bbb57e9c96

* When run as part of the boot loader, it will now adjust the size of partitions
to fit into the session - this should help bug #238 to disappear.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18988 a95241bf-73f2-0310-859d-f6bbb57e9c96

Refactored a bit -- pulled the partition map parser out. Userland add-on disabled for the time being, due to build problems.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9158 a95241bf-73f2-0310-859d-f6bbb57e9c96