freebsd-dev/stand/i386/zfsboot/zfsboot.c

1180 lines
28 KiB
C
Raw Normal View History

Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/*-
* Copyright (c) 1998 Robert Nordier
* All rights reserved.
*
* Redistribution and use in source and binary forms are freely
* permitted provided that the above copyright notice and this
* paragraph and the following disclaimer are duplicated in all
* such forms.
*
* This software is provided "AS IS" and without any express or
* implied warranties, including, without limitation, the implied
* warranties of merchantability and fitness for a particular
* purpose.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "stand.h"
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#include <sys/param.h>
#include <sys/errno.h>
#include <sys/diskmbr.h>
#ifdef GPT
#include <sys/gpt.h>
#endif
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#include <sys/reboot.h>
#include <sys/queue.h>
#include <machine/bootinfo.h>
#include <machine/elf.h>
#include <machine/pc/bios.h>
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#include <stdarg.h>
#include <stddef.h>
#include <a.out.h>
#include <btxv86.h>
#include "lib.h"
- Split code shared by almost any boot loader into separate files and clean up most layering violations: sys/boot/i386/common/rbx.h: RBX_* defines OPT_SET() OPT_CHECK() sys/boot/common/util.[ch]: memcpy() memset() memcmp() bcpy() bzero() bcmp() strcmp() strncmp() [new] strcpy() strcat() strchr() strlen() printf() sys/boot/i386/common/cons.[ch]: ioctrl putc() xputc() putchar() getc() xgetc() keyhit() [now takes number of seconds as an argument] getstr() sys/boot/i386/common/drv.[ch]: struct dsk drvread() drvwrite() [new] drvsize() [new] sys/boot/common/crc32.[ch] [new] sys/boot/common/gpt.[ch] [new] - Teach gptboot and gptzfsboot about new files. I haven't touched the rest, but there is still a lot of code duplication to be removed. - Implement full GPT support. Currently we just read primary header and partition table and don't care about checksums, etc. After this change we verify checksums of primary header and primary partition table and if there is a problem we fall back to backup header and backup partition table. - Clean up most messages to use prefix of boot program, so in case of an error we know where the error comes from, eg.: gptboot: unable to read primary GPT header - If we can't boot, print boot prompt only once and not every five seconds. - Honour newly added GPT attributes: bootme - this is bootable partition bootonce - try to boot from this partition only once bootfailed - we failed to boot from this partition - Change boot order of gptboot to the following: 1. Try to boot from all the partitions that have both 'bootme' and 'bootonce' attributes one by one. 2. Try to boot from all the partitions that have only 'bootme' attribute one by one. 3. If there are no partitions with 'bootme' attribute, boot from the first UFS partition. - The 'bootonce' functionality is implemented in the following way: 1. Walk through all the partitions and when 'bootonce' attribute is found without 'bootme' attribute, remove 'bootonce' attribute and set 'bootfailed' attribute. 'bootonce' attribute alone means that we tried to boot from this partition, but boot failed after leaving gptboot and machine was restarted. 2. Find partition with both 'bootme' and 'bootonce' attributes. 3. Remove 'bootme' attribute. 4. Try to execute /boot/loader or /boot/kernel/kernel from that partition. If succeeded we stop here. 5. If execution failed, remove 'bootonce' and set 'bootfailed'. 6. Go to 2. If whole boot succeeded there is new /etc/rc.d/gptboot script coming that will log all partitions that we failed to boot from (the ones with 'bootfailed' attribute) and will remove this attribute. It will also find partition with 'bootonce' attribute - this is the partition we booted from successfully. The script will log success and remove the attribute. All the GPT updates we do here goes to both primary and backup GPT if they are valid. We don't touch headers or partition tables when checksum doesn't match. Reviewed by: arch (Message-ID: <20100917234542.GE1902@garage.freebsd.pl>) Obtained from: Wheel Systems Sp. z o.o. http://www.wheelsystems.com MFC after: 2 weeks
2010-09-24 19:49:12 +00:00
#include "rbx.h"
#include "drv.h"
#include "edd.h"
- Split code shared by almost any boot loader into separate files and clean up most layering violations: sys/boot/i386/common/rbx.h: RBX_* defines OPT_SET() OPT_CHECK() sys/boot/common/util.[ch]: memcpy() memset() memcmp() bcpy() bzero() bcmp() strcmp() strncmp() [new] strcpy() strcat() strchr() strlen() printf() sys/boot/i386/common/cons.[ch]: ioctrl putc() xputc() putchar() getc() xgetc() keyhit() [now takes number of seconds as an argument] getstr() sys/boot/i386/common/drv.[ch]: struct dsk drvread() drvwrite() [new] drvsize() [new] sys/boot/common/crc32.[ch] [new] sys/boot/common/gpt.[ch] [new] - Teach gptboot and gptzfsboot about new files. I haven't touched the rest, but there is still a lot of code duplication to be removed. - Implement full GPT support. Currently we just read primary header and partition table and don't care about checksums, etc. After this change we verify checksums of primary header and primary partition table and if there is a problem we fall back to backup header and backup partition table. - Clean up most messages to use prefix of boot program, so in case of an error we know where the error comes from, eg.: gptboot: unable to read primary GPT header - If we can't boot, print boot prompt only once and not every five seconds. - Honour newly added GPT attributes: bootme - this is bootable partition bootonce - try to boot from this partition only once bootfailed - we failed to boot from this partition - Change boot order of gptboot to the following: 1. Try to boot from all the partitions that have both 'bootme' and 'bootonce' attributes one by one. 2. Try to boot from all the partitions that have only 'bootme' attribute one by one. 3. If there are no partitions with 'bootme' attribute, boot from the first UFS partition. - The 'bootonce' functionality is implemented in the following way: 1. Walk through all the partitions and when 'bootonce' attribute is found without 'bootme' attribute, remove 'bootonce' attribute and set 'bootfailed' attribute. 'bootonce' attribute alone means that we tried to boot from this partition, but boot failed after leaving gptboot and machine was restarted. 2. Find partition with both 'bootme' and 'bootonce' attributes. 3. Remove 'bootme' attribute. 4. Try to execute /boot/loader or /boot/kernel/kernel from that partition. If succeeded we stop here. 5. If execution failed, remove 'bootonce' and set 'bootfailed'. 6. Go to 2. If whole boot succeeded there is new /etc/rc.d/gptboot script coming that will log all partitions that we failed to boot from (the ones with 'bootfailed' attribute) and will remove this attribute. It will also find partition with 'bootonce' attribute - this is the partition we booted from successfully. The script will log success and remove the attribute. All the GPT updates we do here goes to both primary and backup GPT if they are valid. We don't touch headers or partition tables when checksum doesn't match. Reviewed by: arch (Message-ID: <20100917234542.GE1902@garage.freebsd.pl>) Obtained from: Wheel Systems Sp. z o.o. http://www.wheelsystems.com MFC after: 2 weeks
2010-09-24 19:49:12 +00:00
#include "cons.h"
#include "bootargs.h"
#include "paths.h"
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
#include "libzfs.h"
#define ARGS 0x900
#define NOPT 14
#define NDEV 3
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#define BIOS_NUMDRIVES 0x475
#define DRV_HARD 0x80
#define DRV_MASK 0x7f
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#define TYPE_AD 0
#define TYPE_DA 1
#define TYPE_MAXHARD TYPE_DA
#define TYPE_FD 2
#define DEV_GELIBOOT_BSIZE 4096
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
extern uint32_t _end;
#ifdef GPT
static const uuid_t freebsd_zfs_uuid = GPT_ENT_TYPE_FREEBSD_ZFS;
#endif
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static const char optstr[NOPT] = "DhaCcdgmnpqrsv"; /* Also 'P', 'S' */
static const unsigned char flags[NOPT] = {
RBX_DUAL,
RBX_SERIAL,
RBX_ASKNAME,
RBX_CDROM,
RBX_CONFIG,
RBX_KDB,
RBX_GDB,
RBX_MUTE,
RBX_NOINTR,
RBX_PAUSE,
RBX_QUIET,
RBX_DFLTROOT,
RBX_SINGLE,
RBX_VERBOSE
};
- Split code shared by almost any boot loader into separate files and clean up most layering violations: sys/boot/i386/common/rbx.h: RBX_* defines OPT_SET() OPT_CHECK() sys/boot/common/util.[ch]: memcpy() memset() memcmp() bcpy() bzero() bcmp() strcmp() strncmp() [new] strcpy() strcat() strchr() strlen() printf() sys/boot/i386/common/cons.[ch]: ioctrl putc() xputc() putchar() getc() xgetc() keyhit() [now takes number of seconds as an argument] getstr() sys/boot/i386/common/drv.[ch]: struct dsk drvread() drvwrite() [new] drvsize() [new] sys/boot/common/crc32.[ch] [new] sys/boot/common/gpt.[ch] [new] - Teach gptboot and gptzfsboot about new files. I haven't touched the rest, but there is still a lot of code duplication to be removed. - Implement full GPT support. Currently we just read primary header and partition table and don't care about checksums, etc. After this change we verify checksums of primary header and primary partition table and if there is a problem we fall back to backup header and backup partition table. - Clean up most messages to use prefix of boot program, so in case of an error we know where the error comes from, eg.: gptboot: unable to read primary GPT header - If we can't boot, print boot prompt only once and not every five seconds. - Honour newly added GPT attributes: bootme - this is bootable partition bootonce - try to boot from this partition only once bootfailed - we failed to boot from this partition - Change boot order of gptboot to the following: 1. Try to boot from all the partitions that have both 'bootme' and 'bootonce' attributes one by one. 2. Try to boot from all the partitions that have only 'bootme' attribute one by one. 3. If there are no partitions with 'bootme' attribute, boot from the first UFS partition. - The 'bootonce' functionality is implemented in the following way: 1. Walk through all the partitions and when 'bootonce' attribute is found without 'bootme' attribute, remove 'bootonce' attribute and set 'bootfailed' attribute. 'bootonce' attribute alone means that we tried to boot from this partition, but boot failed after leaving gptboot and machine was restarted. 2. Find partition with both 'bootme' and 'bootonce' attributes. 3. Remove 'bootme' attribute. 4. Try to execute /boot/loader or /boot/kernel/kernel from that partition. If succeeded we stop here. 5. If execution failed, remove 'bootonce' and set 'bootfailed'. 6. Go to 2. If whole boot succeeded there is new /etc/rc.d/gptboot script coming that will log all partitions that we failed to boot from (the ones with 'bootfailed' attribute) and will remove this attribute. It will also find partition with 'bootonce' attribute - this is the partition we booted from successfully. The script will log success and remove the attribute. All the GPT updates we do here goes to both primary and backup GPT if they are valid. We don't touch headers or partition tables when checksum doesn't match. Reviewed by: arch (Message-ID: <20100917234542.GE1902@garage.freebsd.pl>) Obtained from: Wheel Systems Sp. z o.o. http://www.wheelsystems.com MFC after: 2 weeks
2010-09-24 19:49:12 +00:00
uint32_t opts;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/*
* Paths to try loading before falling back to the boot2 prompt.
*
* /boot/zfsloader must be tried before /boot/loader in order to remain
* backward compatible with ZFS boot environments where /boot/loader exists
* but does not have ZFS support, which was the case before FreeBSD 12.
*
* If no loader is found, try to load a kernel directly instead.
*/
static const struct string {
const char *p;
size_t len;
} loadpath[] = {
{ PATH_LOADER_ZFS, sizeof(PATH_LOADER_ZFS) },
{ PATH_LOADER, sizeof(PATH_LOADER) },
{ PATH_KERNEL, sizeof(PATH_KERNEL) },
};
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static const unsigned char dev_maj[NDEV] = {30, 4, 2};
static char cmd[512];
static char cmddup[512];
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static char kname[1024];
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
static char rootname[256];
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static int comspeed = SIOSPD;
static struct bootinfo bootinfo;
static uint32_t bootdev;
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
static struct zfs_boot_args zfsargs;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
vm_offset_t high_heap_base;
uint32_t bios_basemem, bios_extmem, high_heap_size;
static struct bios_smap smap;
/*
* The minimum amount of memory to reserve in bios_extmem for the heap.
*/
#define HEAP_MIN (64 * 1024 * 1024)
static char *heap_next;
static char *heap_end;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/* Buffers that must not span a 64k boundary. */
#define READ_BUF_SIZE 8192
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
struct dmadat {
char rdbuf[READ_BUF_SIZE]; /* for reading large things */
char secbuf[READ_BUF_SIZE]; /* for MBR/disklabel */
};
static struct dmadat *dmadat;
void exit(int);
void reboot(void);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static void load(void);
static int parse_cmd(void);
static void bios_getmem(void);
int main(void);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#ifdef LOADER_GELI_SUPPORT
#include "geliboot.h"
static char gelipw[GELI_PW_MAXLEN];
#endif
struct zfsdsk {
struct dsk dsk;
#ifdef LOADER_GELI_SUPPORT
struct geli_dev *gdev;
#endif
};
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
#include "zfsimpl.c"
/*
* Read from a dnode (which must be from a ZPL filesystem).
*/
static int
zfs_read(spa_t *spa, const dnode_phys_t *dnode, off_t *offp, void *start, size_t size)
{
const znode_phys_t *zp = (const znode_phys_t *) dnode->dn_bonus;
size_t n;
int rc;
n = size;
if (*offp + n > zp->zp_size)
n = zp->zp_size - *offp;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
rc = dnode_read(spa, dnode, *offp, start, n);
if (rc)
return (-1);
*offp += n;
return (n);
}
/*
* Current ZFS pool
*/
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
static spa_t *spa;
static spa_t *primary_spa;
static vdev_t *primary_vdev;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/*
* A wrapper for dskread that doesn't have to worry about whether the
* buffer pointer crosses a 64k boundary.
*/
static int
vdev_read(void *xvdev, void *priv, off_t off, void *buf, size_t bytes)
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
{
char *p;
daddr_t lba, alignlba;
off_t diff;
unsigned int nb, alignnb;
struct zfsdsk *zdsk = (struct zfsdsk *) priv;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
if ((off & (DEV_BSIZE - 1)) || (bytes & (DEV_BSIZE - 1)))
return -1;
p = buf;
lba = off / DEV_BSIZE;
lba += zdsk->dsk.start;
/*
* Align reads to 4k else 4k sector GELIs will not decrypt.
* Round LBA down to nearest multiple of DEV_GELIBOOT_BSIZE bytes.
*/
alignlba = rounddown2(off, DEV_GELIBOOT_BSIZE) / DEV_BSIZE;
/*
* The read must be aligned to DEV_GELIBOOT_BSIZE bytes relative to the
* start of the GELI partition, not the start of the actual disk.
*/
alignlba += zdsk->dsk.start;
diff = (lba - alignlba) * DEV_BSIZE;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
while (bytes > 0) {
nb = bytes / DEV_BSIZE;
/*
* Ensure that the read size plus the leading offset does not
* exceed the size of the read buffer.
*/
if (nb > (READ_BUF_SIZE - diff) / DEV_BSIZE)
nb = (READ_BUF_SIZE - diff) / DEV_BSIZE;
/*
* Round the number of blocks to read up to the nearest multiple
* of DEV_GELIBOOT_BSIZE.
*/
alignnb = roundup2(nb * DEV_BSIZE + diff, DEV_GELIBOOT_BSIZE)
/ DEV_BSIZE;
if (zdsk->dsk.size > 0 && alignlba + alignnb >
zdsk->dsk.size + zdsk->dsk.start) {
printf("Shortening read at %lld from %d to %lld\n",
alignlba, alignnb,
(zdsk->dsk.size + zdsk->dsk.start) - alignlba);
alignnb = (zdsk->dsk.size + zdsk->dsk.start) - alignlba;
}
if (drvread(&zdsk->dsk, dmadat->rdbuf, alignlba, alignnb))
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
return -1;
#ifdef LOADER_GELI_SUPPORT
/* decrypt */
if (zdsk->gdev != NULL) {
if (geli_read(zdsk->gdev, ((alignlba - zdsk->dsk.start) *
DEV_BSIZE), dmadat->rdbuf, alignnb * DEV_BSIZE))
return (-1);
}
#endif
memcpy(p, dmadat->rdbuf + diff, nb * DEV_BSIZE);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
p += nb * DEV_BSIZE;
lba += nb;
alignlba += alignnb;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
bytes -= nb * DEV_BSIZE;
/* Don't need the leading offset after the first block. */
diff = 0;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
return 0;
}
/* Match the signature exactly due to signature madness */
static int
vdev_read2(vdev_t *vdev, void *priv, off_t off, void *buf, size_t bytes)
{
return vdev_read(vdev, priv, off, buf, bytes);
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static int
vdev_write(vdev_t *vdev, void *priv, off_t off, void *buf, size_t bytes)
{
char *p;
daddr_t lba;
unsigned int nb;
struct zfsdsk *zdsk = (struct zfsdsk *) priv;
if ((off & (DEV_BSIZE - 1)) || (bytes & (DEV_BSIZE - 1)))
return -1;
p = buf;
lba = off / DEV_BSIZE;
lba += zdsk->dsk.start;
while (bytes > 0) {
nb = bytes / DEV_BSIZE;
if (nb > READ_BUF_SIZE / DEV_BSIZE)
nb = READ_BUF_SIZE / DEV_BSIZE;
memcpy(dmadat->rdbuf, p, nb * DEV_BSIZE);
if (drvwrite(&zdsk->dsk, dmadat->rdbuf, lba, nb))
return -1;
p += nb * DEV_BSIZE;
lba += nb;
bytes -= nb * DEV_BSIZE;
}
return 0;
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static int
xfsread(const dnode_phys_t *dnode, off_t *offp, void *buf, size_t nbyte)
{
if ((size_t)zfs_read(spa, dnode, offp, buf, nbyte) != nbyte) {
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
printf("Invalid format\n");
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
return -1;
}
return 0;
}
/*
* Read Pad2 (formerly "Boot Block Header") area of the first
* vdev label of the given vdev.
*/
static int
vdev_read_pad2(vdev_t *vdev, char *buf, size_t size)
{
blkptr_t bp;
char *tmp = zap_scratch;
off_t off = offsetof(vdev_label_t, vl_pad2);
if (size > VDEV_PAD_SIZE)
size = VDEV_PAD_SIZE;
BP_ZERO(&bp);
BP_SET_LSIZE(&bp, VDEV_PAD_SIZE);
BP_SET_PSIZE(&bp, VDEV_PAD_SIZE);
BP_SET_CHECKSUM(&bp, ZIO_CHECKSUM_LABEL);
BP_SET_COMPRESS(&bp, ZIO_COMPRESS_OFF);
DVA_SET_OFFSET(BP_IDENTITY(&bp), off);
if (vdev_read_phys(vdev, &bp, tmp, off, 0))
return (EIO);
memcpy(buf, tmp, size);
return (0);
}
static int
vdev_clear_pad2(vdev_t *vdev)
{
char *zeroes = zap_scratch;
uint64_t *end;
off_t off = offsetof(vdev_label_t, vl_pad2);
memset(zeroes, 0, VDEV_PAD_SIZE);
end = (uint64_t *)(zeroes + VDEV_PAD_SIZE);
/* ZIO_CHECKSUM_LABEL magic and pre-calcualted checksum for all zeros */
end[-5] = 0x0210da7ab10c7a11;
end[-4] = 0x97f48f807f6e2a3f;
end[-3] = 0xaf909f1658aacefc;
end[-2] = 0xcbd1ea57ff6db48b;
end[-1] = 0x6ec692db0d465fab;
if (vdev_write(vdev, vdev->v_read_priv, off, zeroes, VDEV_PAD_SIZE))
return (EIO);
return (0);
}
static void
bios_getmem(void)
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
{
uint64_t size;
/* Parse system memory map */
v86.ebx = 0;
do {
v86.ctl = V86_FLAGS;
v86.addr = 0x15; /* int 0x15 function 0xe820*/
v86.eax = 0xe820;
v86.ecx = sizeof(struct bios_smap);
v86.edx = SMAP_SIG;
v86.es = VTOPSEG(&smap);
v86.edi = VTOPOFF(&smap);
v86int();
if (V86_CY(v86.efl) || (v86.eax != SMAP_SIG))
break;
/* look for a low-memory segment that's large enough */
if ((smap.type == SMAP_TYPE_MEMORY) && (smap.base == 0) &&
(smap.length >= (512 * 1024)))
bios_basemem = smap.length;
/* look for the first segment in 'extended' memory */
if ((smap.type == SMAP_TYPE_MEMORY) && (smap.base == 0x100000)) {
bios_extmem = smap.length;
}
/*
* Look for the largest segment in 'extended' memory beyond
* 1MB but below 4GB.
*/
if ((smap.type == SMAP_TYPE_MEMORY) && (smap.base > 0x100000) &&
(smap.base < 0x100000000ull)) {
size = smap.length;
/*
* If this segment crosses the 4GB boundary, truncate it.
*/
if (smap.base + size > 0x100000000ull)
size = 0x100000000ull - smap.base;
if (size > high_heap_size) {
high_heap_size = size;
high_heap_base = smap.base;
}
}
} while (v86.ebx != 0);
/* Fall back to the old compatibility function for base memory */
if (bios_basemem == 0) {
v86.ctl = 0;
v86.addr = 0x12; /* int 0x12 */
v86int();
bios_basemem = (v86.eax & 0xffff) * 1024;
}
/* Fall back through several compatibility functions for extended memory */
if (bios_extmem == 0) {
v86.ctl = V86_FLAGS;
v86.addr = 0x15; /* int 0x15 function 0xe801*/
v86.eax = 0xe801;
v86int();
if (!V86_CY(v86.efl)) {
bios_extmem = ((v86.ecx & 0xffff) + ((v86.edx & 0xffff) * 64)) * 1024;
}
}
if (bios_extmem == 0) {
v86.ctl = 0;
v86.addr = 0x15; /* int 0x15 function 0x88*/
v86.eax = 0x8800;
v86int();
bios_extmem = (v86.eax & 0xffff) * 1024;
}
/*
* If we have extended memory and did not find a suitable heap
* region in the SMAP, use the last 3MB of 'extended' memory as a
* high heap candidate.
*/
if (bios_extmem >= HEAP_MIN && high_heap_size < HEAP_MIN) {
high_heap_size = HEAP_MIN;
high_heap_base = bios_extmem + 0x100000 - HEAP_MIN;
}
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/*
* Try to detect a device supported by the legacy int13 BIOS
*/
static int
int13probe(int drive)
{
v86.ctl = V86_FLAGS;
v86.addr = 0x13;
v86.eax = 0x800;
v86.edx = drive;
v86int();
if (!V86_CY(v86.efl) && /* carry clear */
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
((v86.edx & 0xff) != (drive & DRV_MASK))) { /* unit # OK */
if ((v86.ecx & 0x3f) == 0) { /* absurd sector size */
return(0); /* skip device */
}
return (1);
}
return(0);
}
/*
* We call this when we find a ZFS vdev - ZFS consumes the dsk
* structure so we must make a new one.
*/
static struct zfsdsk *
copy_dsk(struct zfsdsk *zdsk)
{
struct zfsdsk *newdsk;
newdsk = malloc(sizeof(struct zfsdsk));
*newdsk = *zdsk;
return (newdsk);
}
/*
* Get disk size from GPT.
*/
static uint64_t
drvsize_gpt(struct dsk *dskp)
{
#ifdef GPT
struct gpt_hdr hdr;
char *sec;
sec = dmadat->secbuf;
if (drvread(dskp, sec, 1, 1))
return (0);
memcpy(&hdr, sec, sizeof(hdr));
if (memcmp(hdr.hdr_sig, GPT_HDR_SIG, sizeof(hdr.hdr_sig)) != 0 ||
hdr.hdr_lba_self != 1 || hdr.hdr_revision < 0x00010000 ||
hdr.hdr_entsz < sizeof(struct gpt_ent) ||
DEV_BSIZE % hdr.hdr_entsz != 0) {
return (0);
}
return (hdr.hdr_lba_alt + 1);
#else
return (0);
#endif
}
/*
* Get disk size from eax=0x800 and 0x4800. We need to probe both
* because 0x4800 may not be available and we would like to get more
* or less correct disk size - if it is possible at all.
* Note we do not really want to touch drv.c because that code is shared
* with boot2 and we can not afford to grow that code.
*/
static uint64_t
drvsize_ext(struct zfsdsk *zdsk)
{
struct dsk *dskp;
uint64_t size, tmp;
int cyl, hds, sec;
dskp = &zdsk->dsk;
/* Try to read disk size from GPT */
size = drvsize_gpt(dskp);
if (size != 0)
return (size);
v86.ctl = V86_FLAGS;
v86.addr = 0x13;
v86.eax = 0x800;
v86.edx = dskp->drive;
v86int();
/* Don't error out if we get bad sector number, try EDD as well */
if (V86_CY(v86.efl) || /* carry set */
(v86.edx & 0xff) <= (unsigned)(dskp->drive & 0x7f)) /* unit # bad */
return (0);
cyl = ((v86.ecx & 0xc0) << 2) + ((v86.ecx & 0xff00) >> 8) + 1;
/* Convert max head # -> # of heads */
hds = ((v86.edx & 0xff00) >> 8) + 1;
sec = v86.ecx & 0x3f;
size = (uint64_t)cyl * hds * sec;
/* Determine if we can use EDD with this device. */
v86.ctl = V86_FLAGS;
v86.addr = 0x13;
v86.eax = 0x4100;
v86.edx = dskp->drive;
v86.ebx = 0x55aa;
v86int();
if (V86_CY(v86.efl) || /* carry set */
(v86.ebx & 0xffff) != 0xaa55 || /* signature */
(v86.ecx & EDD_INTERFACE_FIXED_DISK) == 0)
return (size);
tmp = drvsize(dskp);
if (tmp > size)
size = tmp;
return (size);
}
/*
* The "layered" ioctl to read disk/partition size. Unfortunately
* the zfsboot case is hardest, because we do not have full software
* stack available, so we need to do some manual work here.
*/
uint64_t
ldi_get_size(void *priv)
{
struct zfsdsk *zdsk = priv;
uint64_t size = zdsk->dsk.size;
if (zdsk->dsk.start == 0)
size = drvsize_ext(zdsk);
return (size * DEV_BSIZE);
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static void
probe_drive(struct zfsdsk *zdsk)
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
{
#ifdef GPT
struct gpt_hdr hdr;
struct gpt_ent *ent;
unsigned part, entries_per_sec;
daddr_t slba;
#endif
#if defined(GPT) || defined(LOADER_GELI_SUPPORT)
daddr_t elba;
#endif
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
struct dos_partition *dp;
char *sec;
unsigned i;
#ifdef LOADER_GELI_SUPPORT
/*
* Taste the disk, if it is GELI encrypted, decrypt it then dig out the
* partition table and probe each slice/partition in turn for a vdev or
* GELI encrypted vdev.
*/
elba = drvsize_ext(zdsk);
if (elba > 0) {
elba--;
}
zdsk->gdev = geli_taste(vdev_read, zdsk, elba, "disk%u:0:");
if ((zdsk->gdev != NULL) && (geli_havekey(zdsk->gdev) == 0))
geli_passphrase(zdsk->gdev, gelipw);
#endif /* LOADER_GELI_SUPPORT */
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
sec = dmadat->secbuf;
zdsk->dsk.start = 0;
#ifdef GPT
/*
* First check for GPT.
*/
if (drvread(&zdsk->dsk, sec, 1, 1)) {
return;
}
memcpy(&hdr, sec, sizeof(hdr));
if (memcmp(hdr.hdr_sig, GPT_HDR_SIG, sizeof(hdr.hdr_sig)) != 0 ||
hdr.hdr_lba_self != 1 || hdr.hdr_revision < 0x00010000 ||
hdr.hdr_entsz < sizeof(*ent) || DEV_BSIZE % hdr.hdr_entsz != 0) {
goto trymbr;
}
/*
* Probe all GPT partitions for the presence of ZFS pools. We
* return the spa_t for the first we find (if requested). This
* will have the effect of booting from the first pool on the
* disk.
*
* If no vdev is found, GELI decrypting the device and try again
*/
entries_per_sec = DEV_BSIZE / hdr.hdr_entsz;
slba = hdr.hdr_lba_table;
elba = slba + hdr.hdr_entries / entries_per_sec;
while (slba < elba) {
zdsk->dsk.start = 0;
if (drvread(&zdsk->dsk, sec, slba, 1))
return;
for (part = 0; part < entries_per_sec; part++) {
ent = (struct gpt_ent *)(sec + part * hdr.hdr_entsz);
if (memcmp(&ent->ent_type, &freebsd_zfs_uuid,
sizeof(uuid_t)) == 0) {
zdsk->dsk.start = ent->ent_lba_start;
zdsk->dsk.size = ent->ent_lba_end - ent->ent_lba_start + 1;
zdsk->dsk.slice = part + 1;
zdsk->dsk.part = 255;
if (vdev_probe(vdev_read2, zdsk, NULL) == 0) {
/*
* This slice had a vdev. We need a new dsk
* structure now since the vdev now owns this one.
*/
zdsk = copy_dsk(zdsk);
}
#ifdef LOADER_GELI_SUPPORT
else if ((zdsk->gdev = geli_taste(vdev_read, zdsk,
ent->ent_lba_end - ent->ent_lba_start, "disk%up%u:",
zdsk->dsk.unit, zdsk->dsk.slice)) != NULL) {
if (geli_havekey(zdsk->gdev) == 0 ||
geli_passphrase(zdsk->gdev, gelipw) == 0) {
/*
* This slice has GELI, check it for ZFS.
*/
if (vdev_probe(vdev_read2, zdsk, NULL) == 0) {
/*
* This slice had a vdev. We need a new dsk
* structure now since the vdev now owns this one.
*/
zdsk = copy_dsk(zdsk);
}
break;
}
}
#endif /* LOADER_GELI_SUPPORT */
}
}
slba++;
}
return;
trymbr:
#endif /* GPT */
if (drvread(&zdsk->dsk, sec, DOSBBSECTOR, 1))
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
return;
dp = (void *)(sec + DOSPARTOFF);
for (i = 0; i < NDOSPART; i++) {
if (!dp[i].dp_typ)
continue;
zdsk->dsk.start = dp[i].dp_start;
zdsk->dsk.size = dp[i].dp_size;
zdsk->dsk.slice = i + 1;
if (vdev_probe(vdev_read2, zdsk, NULL) == 0) {
zdsk = copy_dsk(zdsk);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
#ifdef LOADER_GELI_SUPPORT
else if ((zdsk->gdev = geli_taste(vdev_read, zdsk, dp[i].dp_size -
dp[i].dp_start, "disk%us%u:")) != NULL) {
if (geli_havekey(zdsk->gdev) == 0 ||
geli_passphrase(zdsk->gdev, gelipw) == 0) {
/*
* This slice has GELI, check it for ZFS.
*/
if (vdev_probe(vdev_read2, zdsk, NULL) == 0) {
/*
* This slice had a vdev. We need a new dsk
* structure now since the vdev now owns this one.
*/
zdsk = copy_dsk(zdsk);
}
break;
}
}
#endif /* LOADER_GELI_SUPPORT */
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
}
int
main(void)
{
dnode_phys_t dn;
off_t off;
struct zfsdsk *zdsk;
int autoboot, i;
int nextboot;
int rc;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
dmadat = (void *)(roundup2(__base + (int32_t)&_end, 0x10000) - __base);
bios_getmem();
if (high_heap_size > 0) {
heap_end = PTOV(high_heap_base + high_heap_size);
heap_next = PTOV(high_heap_base);
} else {
heap_next = (char *)dmadat + sizeof(*dmadat);
heap_end = (char *)PTOV(bios_basemem);
}
setheap(heap_next, heap_end);
zdsk = calloc(1, sizeof(struct zfsdsk));
zdsk->dsk.drive = *(uint8_t *)PTOV(ARGS);
zdsk->dsk.type = zdsk->dsk.drive & DRV_HARD ? TYPE_AD : TYPE_FD;
zdsk->dsk.unit = zdsk->dsk.drive & DRV_MASK;
zdsk->dsk.slice = *(uint8_t *)PTOV(ARGS + 1) + 1;
zdsk->dsk.part = 0;
zdsk->dsk.start = 0;
zdsk->dsk.size = drvsize_ext(zdsk);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
bootinfo.bi_version = BOOTINFO_VERSION;
bootinfo.bi_size = sizeof(bootinfo);
bootinfo.bi_basemem = bios_basemem / 1024;
bootinfo.bi_extmem = bios_extmem / 1024;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
bootinfo.bi_memsizes_valid++;
bootinfo.bi_bios_dev = zdsk->dsk.drive;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
bootdev = MAKEBOOTDEV(dev_maj[zdsk->dsk.type],
zdsk->dsk.slice, zdsk->dsk.unit, zdsk->dsk.part);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/* Process configuration file */
autoboot = 1;
zfs_init();
/*
* Probe the boot drive first - we will try to boot from whatever
* pool we find on that drive.
*/
probe_drive(zdsk);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/*
* Probe the rest of the drives that the bios knows about. This
* will find any other available pools and it may fill in missing
* vdevs for the boot pool.
*/
#ifndef VIRTUALBOX
for (i = 0; i < *(unsigned char *)PTOV(BIOS_NUMDRIVES); i++)
#else
for (i = 0; i < MAXBDDEV; i++)
#endif
{
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
if ((i | DRV_HARD) == *(uint8_t *)PTOV(ARGS))
continue;
if (!int13probe(i | DRV_HARD))
break;
zdsk = calloc(1, sizeof(struct zfsdsk));
zdsk->dsk.drive = i | DRV_HARD;
zdsk->dsk.type = zdsk->dsk.drive & TYPE_AD;
zdsk->dsk.unit = i;
zdsk->dsk.slice = 0;
zdsk->dsk.part = 0;
zdsk->dsk.start = 0;
zdsk->dsk.size = drvsize_ext(zdsk);
probe_drive(zdsk);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
/*
* The first discovered pool, if any, is the pool.
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
*/
spa = spa_get_primary();
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
if (!spa) {
printf("%s: No ZFS pools located, can't boot\n", BOOTPROG);
for (;;)
;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
primary_spa = spa;
primary_vdev = spa_get_primary_vdev(spa);
nextboot = 0;
rc = vdev_read_pad2(primary_vdev, cmd, sizeof(cmd));
if (vdev_clear_pad2(primary_vdev))
printf("failed to clear pad2 area of primary vdev\n");
if (rc == 0) {
if (*cmd) {
/*
* We could find an old-style ZFS Boot Block header here.
* Simply ignore it.
*/
if (*(uint64_t *)cmd != 0x2f5b007b10c) {
/*
* Note that parse() is destructive to cmd[] and we also want
* to honor RBX_QUIET option that could be present in cmd[].
*/
nextboot = 1;
memcpy(cmddup, cmd, sizeof(cmd));
if (parse_cmd()) {
printf("failed to parse pad2 area of primary vdev\n");
reboot();
}
if (!OPT_CHECK(RBX_QUIET))
printf("zfs nextboot: %s\n", cmddup);
}
/* Do not process this command twice */
*cmd = 0;
}
} else
printf("failed to read pad2 area of primary vdev\n");
/* Mount ZFS only if it's not already mounted via nextboot parsing. */
if (zfsmount.spa == NULL &&
(zfs_spa_init(spa) != 0 || zfs_mount(spa, 0, &zfsmount) != 0)) {
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
printf("%s: failed to mount default pool %s\n",
BOOTPROG, spa->spa_name);
autoboot = 0;
} else if (zfs_lookup(&zfsmount, PATH_CONFIG, &dn) == 0 ||
zfs_lookup(&zfsmount, PATH_DOTCONFIG, &dn) == 0) {
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
off = 0;
zfs_read(spa, &dn, &off, cmd, sizeof(cmd));
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
if (*cmd) {
/*
* Note that parse_cmd() is destructive to cmd[] and we also want
* to honor RBX_QUIET option that could be present in cmd[].
*/
memcpy(cmddup, cmd, sizeof(cmd));
if (parse_cmd())
autoboot = 0;
if (!OPT_CHECK(RBX_QUIET))
printf("%s: %s\n", PATH_CONFIG, cmddup);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/* Do not process this command twice */
*cmd = 0;
}
/* Do not risk waiting at the prompt forever. */
if (nextboot && !autoboot)
reboot();
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
if (autoboot && !*kname) {
/*
* Iterate through the list of loader and kernel paths, trying to load.
* If interrupted by a keypress, or in case of failure, drop the user
* to the boot2 prompt.
*/
for (i = 0; i < nitems(loadpath); i++) {
memcpy(kname, loadpath[i].p, loadpath[i].len);
if (keyhit(3))
break;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
load();
}
}
/* Present the user with the boot2 prompt. */
for (;;) {
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
if (!autoboot || !OPT_CHECK(RBX_QUIET)) {
printf("\nFreeBSD/x86 boot\n");
if (zfs_rlookup(spa, zfsmount.rootobj, rootname) != 0)
printf("Default: %s/<0x%llx>:%s\n"
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
"boot: ",
spa->spa_name, zfsmount.rootobj, kname);
else if (rootname[0] != '\0')
printf("Default: %s/%s:%s\n"
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
"boot: ",
spa->spa_name, rootname, kname);
else
printf("Default: %s:%s\n"
"boot: ",
spa->spa_name, kname);
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
if (ioctrl & IO_SERIAL)
sio_flush();
- Split code shared by almost any boot loader into separate files and clean up most layering violations: sys/boot/i386/common/rbx.h: RBX_* defines OPT_SET() OPT_CHECK() sys/boot/common/util.[ch]: memcpy() memset() memcmp() bcpy() bzero() bcmp() strcmp() strncmp() [new] strcpy() strcat() strchr() strlen() printf() sys/boot/i386/common/cons.[ch]: ioctrl putc() xputc() putchar() getc() xgetc() keyhit() [now takes number of seconds as an argument] getstr() sys/boot/i386/common/drv.[ch]: struct dsk drvread() drvwrite() [new] drvsize() [new] sys/boot/common/crc32.[ch] [new] sys/boot/common/gpt.[ch] [new] - Teach gptboot and gptzfsboot about new files. I haven't touched the rest, but there is still a lot of code duplication to be removed. - Implement full GPT support. Currently we just read primary header and partition table and don't care about checksums, etc. After this change we verify checksums of primary header and primary partition table and if there is a problem we fall back to backup header and backup partition table. - Clean up most messages to use prefix of boot program, so in case of an error we know where the error comes from, eg.: gptboot: unable to read primary GPT header - If we can't boot, print boot prompt only once and not every five seconds. - Honour newly added GPT attributes: bootme - this is bootable partition bootonce - try to boot from this partition only once bootfailed - we failed to boot from this partition - Change boot order of gptboot to the following: 1. Try to boot from all the partitions that have both 'bootme' and 'bootonce' attributes one by one. 2. Try to boot from all the partitions that have only 'bootme' attribute one by one. 3. If there are no partitions with 'bootme' attribute, boot from the first UFS partition. - The 'bootonce' functionality is implemented in the following way: 1. Walk through all the partitions and when 'bootonce' attribute is found without 'bootme' attribute, remove 'bootonce' attribute and set 'bootfailed' attribute. 'bootonce' attribute alone means that we tried to boot from this partition, but boot failed after leaving gptboot and machine was restarted. 2. Find partition with both 'bootme' and 'bootonce' attributes. 3. Remove 'bootme' attribute. 4. Try to execute /boot/loader or /boot/kernel/kernel from that partition. If succeeded we stop here. 5. If execution failed, remove 'bootonce' and set 'bootfailed'. 6. Go to 2. If whole boot succeeded there is new /etc/rc.d/gptboot script coming that will log all partitions that we failed to boot from (the ones with 'bootfailed' attribute) and will remove this attribute. It will also find partition with 'bootonce' attribute - this is the partition we booted from successfully. The script will log success and remove the attribute. All the GPT updates we do here goes to both primary and backup GPT if they are valid. We don't touch headers or partition tables when checksum doesn't match. Reviewed by: arch (Message-ID: <20100917234542.GE1902@garage.freebsd.pl>) Obtained from: Wheel Systems Sp. z o.o. http://www.wheelsystems.com MFC after: 2 weeks
2010-09-24 19:49:12 +00:00
if (!autoboot || keyhit(5))
getstr(cmd, sizeof(cmd));
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
else if (!autoboot || !OPT_CHECK(RBX_QUIET))
putchar('\n');
autoboot = 0;
if (parse_cmd())
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
putchar('\a');
else
load();
}
}
/* XXX - Needed for btxld to link the boot2 binary; do not remove. */
void
exit(int x)
{
__exit(x);
}
void
reboot(void)
{
__exit(0);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
static void
load(void)
{
union {
struct exec ex;
Elf32_Ehdr eh;
} hdr;
static Elf32_Phdr ep[2];
static Elf32_Shdr es[2];
caddr_t p;
dnode_phys_t dn;
off_t off;
uint32_t addr, x;
int fmt, i, j;
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
if (zfs_lookup(&zfsmount, kname, &dn)) {
printf("\nCan't find %s\n", kname);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
return;
}
off = 0;
if (xfsread(&dn, &off, &hdr, sizeof(hdr)))
return;
if (N_GETMAGIC(hdr.ex) == ZMAGIC)
fmt = 0;
else if (IS_ELF(hdr.eh))
fmt = 1;
else {
printf("Invalid %s\n", "format");
return;
}
if (fmt == 0) {
addr = hdr.ex.a_entry & 0xffffff;
p = PTOV(addr);
off = PAGE_SIZE;
if (xfsread(&dn, &off, p, hdr.ex.a_text))
return;
p += roundup2(hdr.ex.a_text, PAGE_SIZE);
if (xfsread(&dn, &off, p, hdr.ex.a_data))
return;
p += hdr.ex.a_data + roundup2(hdr.ex.a_bss, PAGE_SIZE);
bootinfo.bi_symtab = VTOP(p);
memcpy(p, &hdr.ex.a_syms, sizeof(hdr.ex.a_syms));
p += sizeof(hdr.ex.a_syms);
if (hdr.ex.a_syms) {
if (xfsread(&dn, &off, p, hdr.ex.a_syms))
return;
p += hdr.ex.a_syms;
if (xfsread(&dn, &off, p, sizeof(int)))
return;
x = *(uint32_t *)p;
p += sizeof(int);
x -= sizeof(int);
if (xfsread(&dn, &off, p, x))
return;
p += x;
}
} else {
off = hdr.eh.e_phoff;
for (j = i = 0; i < hdr.eh.e_phnum && j < 2; i++) {
if (xfsread(&dn, &off, ep + j, sizeof(ep[0])))
return;
if (ep[j].p_type == PT_LOAD)
j++;
}
for (i = 0; i < 2; i++) {
p = PTOV(ep[i].p_paddr & 0xffffff);
off = ep[i].p_offset;
if (xfsread(&dn, &off, p, ep[i].p_filesz))
return;
}
p += roundup2(ep[1].p_memsz, PAGE_SIZE);
bootinfo.bi_symtab = VTOP(p);
if (hdr.eh.e_shnum == hdr.eh.e_shstrndx + 3) {
off = hdr.eh.e_shoff + sizeof(es[0]) *
(hdr.eh.e_shstrndx + 1);
if (xfsread(&dn, &off, &es, sizeof(es)))
return;
for (i = 0; i < 2; i++) {
memcpy(p, &es[i].sh_size, sizeof(es[i].sh_size));
p += sizeof(es[i].sh_size);
off = es[i].sh_offset;
if (xfsread(&dn, &off, p, es[i].sh_size))
return;
p += es[i].sh_size;
}
}
addr = hdr.eh.e_entry & 0xffffff;
}
bootinfo.bi_esymtab = VTOP(p);
bootinfo.bi_kernelname = VTOP(kname);
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
zfsargs.size = sizeof(zfsargs);
zfsargs.pool = zfsmount.spa->spa_guid;
zfsargs.root = zfsmount.rootobj;
zfsargs.primary_pool = primary_spa->spa_guid;
#ifdef LOADER_GELI_SUPPORT
explicit_bzero(gelipw, sizeof(gelipw));
export_geli_boot_data(&zfsargs.gelidata);
#endif
if (primary_vdev != NULL)
zfsargs.primary_vdev = primary_vdev->v_guid;
else
printf("failed to detect primary vdev\n");
/*
* Note that the zfsargs struct is passed by value, not by pointer. Code in
* btxldr.S copies the values from the entry stack to a fixed location
* within loader(8) at startup due to the presence of KARGS_FLAGS_EXTARG.
*/
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
__exec((caddr_t)addr, RB_BOOTINFO | (opts & RBX_MASK),
bootdev,
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
KARGS_FLAGS_ZFS | KARGS_FLAGS_EXTARG,
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
(uint32_t) spa->spa_guid,
(uint32_t) (spa->spa_guid >> 32),
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
VTOP(&bootinfo),
zfsargs);
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
static int
zfs_mount_ds(char *dsname)
{
uint64_t newroot;
spa_t *newspa;
char *q;
q = strchr(dsname, '/');
if (q)
*q++ = '\0';
newspa = spa_find_by_name(dsname);
if (newspa == NULL) {
printf("\nCan't find ZFS pool %s\n", dsname);
return -1;
}
if (zfs_spa_init(newspa))
return -1;
newroot = 0;
if (q) {
if (zfs_lookup_dataset(newspa, q, &newroot)) {
printf("\nCan't find dataset %s in ZFS pool %s\n",
q, newspa->spa_name);
return -1;
}
}
if (zfs_mount(newspa, newroot, &zfsmount)) {
printf("\nCan't mount ZFS dataset\n");
return -1;
}
spa = newspa;
return (0);
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
static int
parse_cmd(void)
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
{
char *arg = cmd;
char *ep, *p, *q;
const char *cp;
int c, i, j;
while ((c = *arg++)) {
if (c == ' ' || c == '\t' || c == '\n')
continue;
for (p = arg; *p && *p != '\n' && *p != ' ' && *p != '\t'; p++);
ep = p;
if (*p)
*p++ = 0;
if (c == '-') {
while ((c = *arg++)) {
if (c == 'P') {
if (*(uint8_t *)PTOV(0x496) & 0x10) {
cp = "yes";
} else {
opts |= OPT_SET(RBX_DUAL) | OPT_SET(RBX_SERIAL);
cp = "no";
}
printf("Keyboard: %s\n", cp);
continue;
} else if (c == 'S') {
j = 0;
while ((unsigned int)(i = *arg++ - '0') <= 9)
j = j * 10 + i;
if (j > 0 && i == -'0') {
comspeed = j;
break;
}
/* Fall through to error below ('S' not in optstr[]). */
}
for (i = 0; c != optstr[i]; i++)
if (i == NOPT - 1)
return -1;
opts ^= OPT_SET(flags[i]);
}
ioctrl = OPT_CHECK(RBX_DUAL) ? (IO_SERIAL|IO_KEYBOARD) :
OPT_CHECK(RBX_SERIAL) ? IO_SERIAL : IO_KEYBOARD;
if (ioctrl & IO_SERIAL) {
if (sio_init(115200 / comspeed) != 0)
ioctrl &= ~IO_SERIAL;
}
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
} if (c == '?') {
dnode_phys_t dn;
zfsboot/zfsloader: support accessing filesystems within a pool In zfs loader zfs device name format now is "zfs:pool/fs", fully qualified file path is "zfs:pool/fs:/path/to/file" loader allows accessing files from various pools and filesystems as well as changing currdev to a different pool/filesystem. zfsboot accepts kernel/loader name in a format pool:fs:path/to/file or, as before, pool:path/to/file; in the latter case a default filesystem is used (pool root or bootfs). zfsboot passes guids of the selected pool and dataset to zfsloader to be used as its defaults. zfs support should be architecture independent and is provided in a separate library, but architectures wishing to use this zfs support still have to provide some glue code and their devdesc should be compatible with zfs_devdesc. arch_zfs_probe method is used to discover all disk devices that may be part of ZFS pool(s). libi386 unconditionally includes zfs support, but some zfs-specific functions are stubbed out as weak symbols. The strong definitions are provided in libzfsboot. This change mean that the size of i386_devspec becomes larger to match zfs_devspec. Backward-compatibility shims are provided for recently added sparc64 zfs boot support. Currently that architecture still works the old way and does not support the new features. TODO: - clear up pool root filesystem vs pool bootfs filesystem distinction - update sparc64 support - set vfs.root.mountfrom based on currdev (for zfs) Mid-future TODO: - loader sub-menu for selecting alternative boot environment Distant future TODO: - support accessing snapshots, using a snapshot as readonly root Reviewed by: marius (sparc64), Gavin Mu <gavin.mu@gmail.com> (sparc64) Tested by: Florian Wagner <florian@wagner-flo.net> (x86), marius (sparc64) No objections: fs@, hackers@ MFC after: 1 month
2012-05-12 09:03:30 +00:00
if (zfs_lookup(&zfsmount, arg, &dn) == 0) {
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
zap_list(spa, &dn);
}
return -1;
} else {
arg--;
/*
* Report pool status if the comment is 'status'. Lets
* hope no-one wants to load /status as a kernel.
*/
if (!strcmp(arg, "status")) {
spa_all_status();
return -1;
}
/*
* If there is "zfs:" prefix simply ignore it.
*/
if (strncmp(arg, "zfs:", 4) == 0)
arg += 4;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
/*
* If there is a colon, switch pools.
*/
q = strchr(arg, ':');
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
if (q) {
*q++ = '\0';
if (zfs_mount_ds(arg) != 0)
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
return -1;
arg = q;
Update ZFS from version 6 to 13 and bring some FreeBSD-specific changes. This bring huge amount of changes, I'll enumerate only user-visible changes: - Delegated Administration Allows regular users to perform ZFS operations, like file system creation, snapshot creation, etc. - L2ARC Level 2 cache for ZFS - allows to use additional disks for cache. Huge performance improvements mostly for random read of mostly static content. - slog Allow to use additional disks for ZFS Intent Log to speed up operations like fsync(2). - vfs.zfs.super_owner Allows regular users to perform privileged operations on files stored on ZFS file systems owned by him. Very careful with this one. - chflags(2) Not all the flags are supported. This still needs work. - ZFSBoot Support to boot off of ZFS pool. Not finished, AFAIK. Submitted by: dfr - Snapshot properties - New failure modes Before if write requested failed, system paniced. Now one can select from one of three failure modes: - panic - panic on write error - wait - wait for disk to reappear - continue - serve read requests if possible, block write requests - Refquota, refreservation properties Just quota and reservation properties, but don't count space consumed by children file systems, clones and snapshots. - Sparse volumes ZVOLs that don't reserve space in the pool. - External attributes Compatible with extattr(2). - NFSv4-ACLs Not sure about the status, might not be complete yet. Submitted by: trasz - Creation-time properties - Regression tests for zpool(8) command. Obtained from: OpenSolaris
2008-11-17 20:49:29 +00:00
}
if ((i = ep - arg)) {
if ((size_t)i >= sizeof(kname))
return -1;
memcpy(kname, arg, i + 1);
}
}
arg = p;
}
return 0;
}