summaryrefslogtreecommitdiffstats
path: root/perl-install/fs
Commit message (Collapse)AuthorAgeFilesLines
* Force reinitialisation of partition suggestions after wiping disk.Martin Whitaker2017-04-041-1/+1
| | | | | | | | When the "Use entire disk" partitioning solution is chosen in a UEFI install, any existing ESP will be deleted, so we must force the partition suggestions to be updated to ensure we create a new one. (this matches what is done in the diskdrake auto-allocate routine)
* Speed up proc_partitions::compare() and avoid unwanted udev events.Martin Whitaker2017-04-041-3/+3
| | | | | | | | The compare function doesn't look at the filesystem type, so there's no need to probe it. This avoids unwanted udev events when we are checking that the kernel has been informed of partition table changes.
* Don't require a BIOS boot partition when performing an upgrade (mga#20338)Martin Whitaker2017-04-041-1/+3
| | | | | | | | If we are upgrading a system on a GPT disk and it doesn't currently have a BIOS boot partition, the old system must be using a legacy bootloader (e.g. grub). We support the continued use of legacy bootloaders in an upgrade, so there is no need to insist on a BIOS boot partition in this case.
* Allow a BIOS boot partition to be located on any disk (mga#20161).Martin Whitaker2017-02-252-19/+18
| | | | | | | This relaxes the rules in fs::any::is_boot_bios_part_needed to allow the BIOS boot partition to be located on any disk, not just the disk containing the root partition. It also ensures that a BIOS boot partition is never forcibly required in a system with non-GPT disks.
* Set fs_type to '' for BIOS boot and Empty partitions.Martin Whitaker2017-02-251-0/+1
| | | | | This prevents misreporting of the partition type in diskdrake if they located on top of an old partition that contained a valid file system.
* Don't suggest mountpoint for ESP when doing a legacy boot install (mga#20164).Martin Whitaker2017-02-251-1/+1
| | | | | | When doing a UEFI install, we add a fstab entry to mount the ESP on /boot/EFI. This is neither required nor desirable when doing a legacy boot install, even if an ESP is present on the disk.
* Don't treat Empty or BIOS_GRUB partitions as free space (mga#20074).Martin Whitaker2017-02-251-1/+1
| | | | | | | The subroutine isEmpty() is used to identify free space on the disk (not DOS "Empty" partitions). Because we use a string to flag Empty and BIOS_GRUB partitions, rather than a numeric value, the numeric equality operator gives false positives.
* Preserve DOS "Empty" partitions instead of treating them as free space ↵Martin Whitaker2017-02-251-1/+1
| | | | | | | | | (mga#20074). To minimise the changes this close to mga6 release (and until we fix the bug in partition auto-allocation that mistakenly creates BIOS boot partitions non-GPT disks), reuse the BIOS_GRUB flag for flagging Empty partitions, as the real partition ID (0x00) is used to flag free space.
* Fix bug in detecting whether a BIOS boot partition is required (mga#19935).Martin Whitaker2017-01-171-1/+1
|
* Skip check for boot partitions when building Live images (mga#19516).Martin Whitaker2016-12-021-1/+3
|
* (isFormatable) introduce itThierry Vignaud2016-11-151-1/+7
| | | | | we could have exported cannotBeMountable() but it's cleaner to export another isFoobar() function
* simplifyThierry Vignaud2016-10-161-1/+1
|
* make warning more explicit about "BBP" purposeRémi Verschelde2016-07-122-2/+2
| | | | (BIOS boot partition, thus fixing mga#18923)
* BIOS boot partition cannot be mountedThierry Vignaud2016-07-111-1/+1
|
* fix inverted testThierry Vignaud2016-07-051-1/+1
|
* factorize bad block checksThierry Vignaud2016-07-042-2/+3
| | | | | also enable bad blocks checking on JFS (was in one place but not in other and indeed mkfs.jfs does support it)
* perl_checker cleanupThierry Vignaud2016-07-041-1/+1
|
* fix being able to create vfat on UEFI (mga#15698)Thierry Vignaud2016-06-301-1/+2
| | | | | | | | | and explain the purpose of the uniq_() call see commit 56b573476e2ea63a509ab2f8ec0683397ae7c85e Rationale: use_pt_type() now returns 0 for common disks (eg: GPT ones) Under GPT+UEFI, we've 2 important "FS" having vfag as the real FS: ESP & FAT32
* keep UUID when formating btrfs (mga#18673)Thierry Vignaud2016-06-241-1/+9
|
* update btrfs status regarding keeping UUIDThierry Vignaud2016-06-241-1/+1
| | | | | | we cannot keep UUID when formating btrfs: - btrfstune -U needs answering "y" - mkfs.btrfs -u fails with: "non unique UUID: XYZ..."
* fix preserving UUID when formating jfs/nilfs/xfsThierry Vignaud2016-06-241-1/+1
|
* simplifyThierry Vignaud2016-06-241-1/+1
|
* fix documentionThierry Vignaud2016-06-231-1/+1
|
* add a failsafe for ESP taggingThierry Vignaud2016-06-181-1/+1
| | | | sometimes we end with bogus 'fat32' fs
* fix inverted test (mga#18656, mga#18704)Thierry Vignaud2016-06-161-1/+1
|
* fix offering to create a GRUB_BIOS partitionThierry Vignaud2016-06-121-1/+1
| | | | ...in custom mode (mga#18656)
* fix partition type testThierry Vignaud2016-06-121-0/+1
| | | | thus guessing better if we need a GRUB_BIOS partition (mga#18656)
* update suggestions if needed after erasing diskThierry Vignaud2016-06-111-0/+3
| | | | | | this is incomplete: we need to reread $all_hds else we won't add a Boot BIOS partition to suggestions if there was already one _prior_ to erase the full disk (mga#18656) ...
* add a GRUB_BIOS partitions if needed (mga#18656)Thierry Vignaud2016-06-112-0/+38
|
* init_efi_suggestions -> init_mntpnt_suggestions()Thierry Vignaud2016-06-111-1/+1
| | | | | also take a $all_hds arg instead of $fstab needed for next commit
* refactor for next commitThierry Vignaud2016-06-111-2/+4
|
* detect GRUB_BIOS partitions (mga#18656)Thierry Vignaud2016-06-111-3/+5
| | | | let's abuse ->{pt_type} for tracking such partitions
* check that ESP mount point actually is an ESPThierry Vignaud2016-02-101-1/+4
| | | | fixing grub2 faillure resulting in draklive-install crash (mga#16246)
* (isESP) check fs too (mga#16246)Thierry Vignaud2016-02-101-1/+1
|
* drop (broken) support for Alpha, IA64 & PPC, SparcThierry Vignaud2016-01-021-1/+1
|
* prevent a crash (mga#15919)Thierry Vignaud2015-12-251-1/+3
|
* Suppress undefined comment in secrets line mga#10540Papoteur2015-12-151-1/+1
|
* only use current disk when auto partitioning, wiping disk, using free space ↵Pascal Terjan2015-06-031-3/+3
| | | | (mga#16055)
* resync with udisk2Thierry Vignaud2015-05-281-1/+4
| | | | split those unknow from udisk2 on next line
* update commentThierry Vignaud2015-05-281-2/+2
|
* fix last commit (spotted by tv), update intaller NEWS tooThomas Backlund2015-05-281-1/+1
|
* ignore special 'Packard_Bell' and 'Push_Button_Reset' partitions (mga#15999)Thomas Backlund2015-05-281-1/+1
|
* fix not always suggesting /boot/ESP mount pointMartin Whitaker2015-05-251-0/+7
| | | | | | | | | | | | | thus fixing mga#15689 for live installs The check for whether an ESP needs to be created or not seems to prevent us to suggest a mount point for the existing one in some cases to be reviewed for mga6... diff --git a/perl-install/fs/partitioning_wizard.pm b/perl-install/fs/partitioning_wizard.pm index f3b9973..b222c07 100644 --- a/perl-install/fs/partitioning_wizard.pm
* perl_checker cleanupThierry Vignaud2015-05-181-1/+1
|
* enables to get raw available space (mga15257)Thierry Vignaud2015-05-181-2/+2
|
* (getNeededMinSpace) split it outThierry Vignaud2015-05-181-3/+9
| | | | needed for mga#15257
* fix regression refusing FAT for ESP (mga#15962)Thierry Vignaud2015-05-171-1/+4
| | | | regression introduced in commit ae5b3ce9845669227ef06c90dfaa1065a7ceebe8
* perl_checker fixThierry Vignaud2015-05-171-1/+1
|
* log which disk we've issue withThierry Vignaud2015-05-171-1/+1
|
* fix a crash after "insufficient space" errorThierry Vignaud2015-05-141-1/+1
| | | | | | | | | | | | | | | | | create_display_box gets called twice. The first time, only the $kind parameter has a value, the second time all the parameters (including $button) have values. On the first time through the partitioning step, the second call to create_display_box sets the selected_for_resize flag for my NTFS partition. When returning to the partitioning step, this flag is still set, so the first call to create_display_box executes the block of code that ends with $button->signal_connect('focus-in-event' => sub { ... }); even though it doesn't have a value for $button. (Martin Whitaker, mga#15919)
generated by cgit v1.2.1 (git 2.21.0) at 2024-06-20 18:28:06 +0000