NetBSD/usr.sbin/sysinst/partitions.c

234 lines
6.4 KiB
C
Raw Normal View History

/* $NetBSD: partitions.c,v 1.13 2021/09/11 20:28:06 andvar Exp $ */
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
/*
2020-11-06 15:23:10 +03:00
* Copyright (c) 2020 The NetBSD Foundation, Inc.
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
2020-11-06 15:23:10 +03:00
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
2020-11-06 15:23:10 +03:00
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
*/
#include "defs.h"
#include "mbr.h"
#include <assert.h>
/*
* A list of partitioning schemes, so we can iterate over everything
* supported (e.g. when partitioning a new disk). NULL terminated.
*/
const struct disk_partitioning_scheme **available_part_schemes;
/*
* The number of valid entries on above list
*/
size_t num_available_part_schemes;
extern const struct disk_partitioning_scheme disklabel_parts;
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
/*
* Generic reader - query a disk device and read all partitions from it.
* disk_size is in units of physical sector size, which is passe as
* bytes_per_sec.
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
*/
struct disk_partitions *
partitions_read_disk(const char *dev, daddr_t disk_size, size_t bytes_per_sec,
bool no_mbr)
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
{
const struct disk_partitioning_scheme **ps;
#ifdef HAVE_MBR
bool mbr_done = false, disklabel_done = false;
#endif
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
if (!available_part_schemes)
return NULL;
for (ps = available_part_schemes; *ps; ps++) {
#ifdef HAVE_MBR
if (!no_mbr && (*ps) == &disklabel_parts && !mbr_done)
continue;
if (no_mbr && (*ps)->name == MSG_parttype_mbr)
continue;
if ((*ps)->name == MSG_parttype_mbr)
mbr_done = true;
if ((*ps)->read_from_disk == disklabel_parts.read_from_disk)
disklabel_done = true;
#endif
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
struct disk_partitions *parts =
(*ps)->read_from_disk(dev, 0, disk_size, bytes_per_sec,
*ps);
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
if (parts)
return parts;
}
#ifdef HAVE_MBR
if (!disklabel_done)
return disklabel_parts.read_from_disk(dev, 0, disk_size,
bytes_per_sec, &disklabel_parts);
#endif
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
return NULL;
}
bool
generic_adapt_foreign_part_info(const struct disk_partitions *myself,
struct disk_part_info *dest,
const struct disk_partitioning_scheme *src_scheme,
const struct disk_part_info *src)
{
*dest = *src;
if (myself->pscheme == src_scheme)
return true; /* no conversion needed */
if (src->nat_type == NULL)
return false;
/* slightly simplistic, enhance when needed */
dest->nat_type = myself->pscheme->get_fs_part_type(
dest->nat_type ? dest->nat_type->generic_ptype : PT_root,
dest->fs_type,
dest->fs_sub_type);
if (dest->nat_type == NULL)
dest->nat_type = myself->pscheme->get_generic_part_type(
src->nat_type->generic_ptype);
if (dest->nat_type == NULL)
dest->nat_type = myself->pscheme->create_unknown_part_type();
if (dest->nat_type == NULL)
dest->nat_type = myself->pscheme->get_generic_part_type(
PT_unknown);
return true;
}
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
/*************** global init ****************************************/
/*
* Helper structure to fill our global list of available partitioning
* schemes.
*/
struct part_scheme_desc {
bool (*is_available)(void);
const struct disk_partitioning_scheme *ps;
};
#ifdef HAVE_GPT
bool gpt_parts_check(void);
extern const struct disk_partitioning_scheme gpt_parts;
#endif
#ifdef HAVE_MBR
extern const struct disk_partitioning_scheme mbr_parts;
#endif
#if RAW_PART == 3
static struct disk_partitioning_scheme only_disklabel_parts;
/*
* If not overridden by MD code, we can not boot from plain
* disklabel disks (w/o MBR).
*/
static bool have_only_disklabel_boot_support(const char *disk)
{
#ifdef HAVE_PLAIN_DISKLABEL_BOOT
return HAVE_PLAIN_DISKLABEL_BOOT(disk);
#else
return false;
#endif
}
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
#endif
/*
* One time initialization
*/
void
partitions_init(void)
{
/*
* List of partitioning schemes.
* Order is important, the selection menu is created from start
* to end. Keep good defaults early. Most architectures will
* only offer very few entries.
*/
static const struct part_scheme_desc all_descs[] = {
#if RAW_PART != 3 /* only available as primary on some architectures */
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
{ NULL, &disklabel_parts },
#endif
#ifdef HAVE_GPT
{ gpt_parts_check, &gpt_parts },
#endif
#ifdef HAVE_MBR
{ NULL, &mbr_parts },
#endif
#if RAW_PART == 3 /* "whole disk NetBSD" disklabel variant */
{ NULL, &only_disklabel_parts },
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
#endif
};
size_t i, avail;
const struct disk_partitioning_scheme **out;
bool *is_available;
static const size_t all_cnt = __arraycount(all_descs);
check_available_binaries();
#if RAW_PART == 3
/* generate a variant of disklabel w/o parent scheme */
only_disklabel_parts = disklabel_parts;
only_disklabel_parts.name = MSG_parttype_only_disklabel;
only_disklabel_parts.have_boot_support =
have_only_disklabel_boot_support;
#endif
Rework internal data structures and "interfaces to user interface" functions to get rid of all disklabel assumptions. Previously (even for GPT partitioning) struct disklabel was used, which obviously breaks large disk setups. Also many MD parts and parts of the user interface assumed (a) a struct disklabel is used internally to store partitioning information and (b) partitions are named 'a' ... $MAXPART. Get rid of this and replace it with a quite abstract interface that should be able to deal with all variants in partition storage: - partitions are stored in a (partly abstract) struct disk_partitions and most parts of it are only accessed via accessor functions provided by a "partitioning scheme". - implement partitioning schemes for MBR, disklabel and GPT (with likely RDB [amiga] and Apple Partition Map [mac*] to follow soon) - partitioning schemes may be cascaded, e.g. on x86 when using MBR as "outer partitions", we have disklabel as "inner partitions". - all user interface goes via accessor functions in the partitioning scheme, some of which return pointers to special user interface descriptors (e.g. to allow editing partition flags, which are scheme specific) Overall the user interface changes (in this initial step) are minimal but noticable. A new Anita is needed for automatic test setups - many thanks to Andreas Gustafsson for lots of early testing and a new Anita version, and to Manuel Bouyer for cooperation and tests of the Anita release. This work was sponsored by The NetBSD Foundation, Inc.
2019-06-12 09:20:17 +03:00
is_available = malloc(all_cnt);
for (avail = i = 0; i < all_cnt; i++) {
is_available[i] = all_descs[i].is_available == NULL
|| all_descs[i].is_available();
if (is_available[i])
avail++;
}
if (avail == 0)
return;
num_available_part_schemes = avail;
available_part_schemes = malloc(sizeof(*available_part_schemes)
* (avail+1));
if (available_part_schemes == NULL)
return;
for (out = available_part_schemes, i = 0; i < all_cnt; i++) {
if (!is_available[i])
continue;
*out++ = all_descs[i].ps;
}
*out = NULL;
free(is_available);
}
/*
* Final cleanup
*/
void
partitions_cleanup(void)
{
for (size_t i = 0; i < num_available_part_schemes; i++)
if (available_part_schemes[i]->cleanup != NULL)
available_part_schemes[i]->cleanup();
free(available_part_schemes);
}