A user reported that formatting a whole disk device with a file system
failed like this:
Format /dev/sdd as ext4 (ERROR)
+ calibrate /dev/sdd (SUCCESS)
path: /dev/sdd (device)
start: 0
end: 15633407
size: 15633408 (7.45 GiB)
+ clear old file system signatures in /dev/sdd (SUCCESS)
+ write 512.00 KiB of zeros at byte offset 0 (SUCCESS)
+ write 4.00 KiB of zeros at byte offset 67108864 (SUCCESS)
+ write 512.00 KiB of zeros at byte offset 8003780608 (SUCCESS)
+ write 4.00 KiB of zeros at byte offset 8004239360 (SUCCESS)
+ write 8.00 KiB of zeros at byte offset 8004296704 (SUCCESS)
+ flush operating system cache of /dev/sdd (SUCCESS)
+ create new ext4 file system (ERROR)
+ mkfs.ext4 -F -O ^64bit -L '' '/dev/sdd' (ERROR)
mke2fs 1.44.1 (24-Mar-2018)
/dev/sdd is apparently in use by the system; will not make a filesystem here!
Opening the whole disk block device exclusively causes mkfs.ext4 to
report that error like this:
# python
>>> import os
>>> f = os.open('/dev/sdb',os.O_RDONLY|os.O_EXCL)
>>> ^Z
[1]+ Stopped python
# mkfs.ext4 -F -O ^64bit -L '' '/dev/sdb'
mke2fs 1.42.9 (28-Dec-2013)
/dev/sdb is apparently in use by the system; will not make a filesystem here!
# echo $?
1
I have not been able to reproduce this error, but with debugging and
sleeping in GParted, stracing GParted and using 'udevadm monitor' to
watch udev events the following sequence of events is seen:
gparted |format(partition, operationdetail)
gparted | erase_filesystem_signatures(partition, operationdetail)
gparted | get_device(device_path="/dev/sdb", lp_device, flush=false)
gparted | ped_device_get("/dev/sdb")
libparted | open("/dev/sdb", O_RDONLY) = 11
libparted | close(11)
gparted | ped_device_open(lp_device)
libparted | open("/dev/sdb", O_RDWR) = 11
gparted | ped_device_sync(lp_device)
libparted | ioctl(11, BLKFLSBUF)
gparted | ped_device_close()
libparted | close(11)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| UDEV change /devices/.../sdb (block)
gparted | set_partition_type(partition, operationdetail)
gparted | create_filesystem(partition, operationdetail)
gparted | ext2::create(partition, operationdetail)
gparted | FileSystem::execute_command("mkfs.ext4 -F -O ^64bit -L '' '/dev/sdb')
So it is assumed that the processing of the udev change rule after
closing the block device in erase_filesystem_signatures() overlaps with
the execution mkfs.ext4 and causes the seen error. Fix by waiting for
those udev events to complete as was previously done by commits [1][2]
[3].
Also note that this is specific to creating file systems on and
formatting unpartitioned whole disk devices because set_partition_type()
is a no-operation. Where as on a partitioned device
set_partition_type() calls commit() which already waits for udev rules
to complete [3].
[1] 50c8924a8e4d9cc96a2ea45f13291114402affee
Wait for udev to recreate /dev/PTN entries when querying partition
FSs (!46)
[2] 4f6c312e3bc68cafb5e6035fd4a5b5bbbfcea992
Wait for udev change on /dev/DISK when querying whole device FS
(!46)
[3] 2f53876c0f
Wait for the kernel and udev to settle partitions for a second time
(#790418)
Closes#83 - /dev/sdd is apparently in use by the system; will not make
a filesystem here!
Name the structure member to 'fsname' used to store strings like "ext2"
etc. This is equivalent to what was previously done in this commit:
a9f08ddc7d
Rename local variable to fsname in get_filesystem() (#741430)
Closes!52 - Rename members and variables currently named 'filesystem'
Previously made this change:
175d27c55d
Rename enum FILESYSTEM to FSType
Now complete the renaming exercise of members and variables currently
named 'filesystem'.
Closes!52 - Rename members and variables currently named 'filesystem'
In GParted_Core::set_device_partitions() the partition path is being
queried from libparted. However this is done before the switch
statement on the type of the partition, so is called for all libparted
partition objects including PED_PARTITION_FREESPACE and
PED_PARTITION_METADATA ones. As libparted numbers these partition
objects as -1, it returns paths like "/dev/sda-1".
Additionally when using GParted, with it's default DMRaid handling, on a
dmraid started array this results in paths like
"/dev/mapper/isw_ecccdhhiga_MyArray-1" being passed to
is_dmraid_device() and make_path_dmraid_compatible(). Fortunately
make_path_dmraid_compatible() does nothing and returns the same name.
Call chain looks like:
GParted_Core::set_device_partitions()
get_partition_path(lp_partition)
// where:
// lp_partition->disk->dev->path = "/dev/mapper/isw_ecccdhhiga_MyArray"
// lp_partition->type == PED_PARTITION_FREESPACE |
// PED_PARTITION_METADATA
// ->num == -1
ped_partition_get_path(lp_partition)
return "/dev/mapper/isw_ecccdhhiga_MyArray-1"
dmraid.is_dmraid_supported()
dmraid.is_dmraid_device("/dev/mapper/isw_ecccdhhiga_MyArray-1")
return true
dmraid.make_path_dmraid_compatible("/dev/mapper/isw_ecccdhhiga_MyArray-1")
return "/dev/mapper/isw_ecccdhhiga_MyArray-1"
Fix by moving the get_partition_path() call inside the switch statement
so that it is only called for PED_PARTITION_NORMAL,
PED_PARTITION_LOGICAL and PED_PARTITION_EXTENDED partition types.
Relevant commits:
* 53c49349f7
Simplify logic in set_device_partitions method
* 81986c0990
Ensure partition path name is compatible with dmraid (#622217)
This is not strictly necessary as members are already recognised using
blkid since this commit earlier in the sequence "Recognise ATARAID
members (#75)". However it makes sure active members are recognised
even if blkid is not available and matches how file system detection
queries the SWRaid_Info module.
Closes#75 - Errors with GPT on RAID 0 ATARAID array
This matches how the array device is displayed as the mount point for
mdadm started ATARAID members by "Display array device as mount point of
mdadm started ATARAID members (#75)" earlier in this patchset.
Extend the DMRaid module member cache to save the array device name and
use as needed to display as the mount point.
Closes#75 - Errors with GPT on RAID 0 ATARAID array
Again this is to stop GParted allowing overwrite operations being
performed on an ATARAID member while the array is actively using the
member. This time for dmraid started arrays using the kernel DM (Device
Mapper) driver.
The DMRaid module already uses dmraid to report active array names:
# dmraid -sa -c
isw_ecccdhhiga_MyArray
To find active members in this array, (1) use udev to lookup the kernel
device name:
# udevadm info --query=name /dev/mapper/isw_ecccdhhiga_MyArray
dm-0
(2) list the member names exposed by the kernel DM driver through the
/sys file system.
# ls /sys/block/dm-0/slaves
sdc sdd
# ls -l /sys/block/dm-0/slaves
lrwxrwxrwx 1 root root 0 Nov 24 09:52 sdc -> ../../../../pci0000:00/0000:00:0d.0/ata3/host2/target2:0:0/2:0:0:0/block/sdc
lrwxrwxrwx 1 root root 0 Nov 24 09:52 sdc -> ../../../../pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdd
Closes#75 - Errors with GPT on RAID 0 ATARAID array
When an ATARAID member is inactive allow basic supported actions of
copy and move to be performed like with other recognised but only basic
supported types.
Closes#75 - Errors with GPT on RAID 0 ATARAID array
Since earlier commit "Display array device as mount point of mdadm
started ATARAID members (#75)" GParted allows attempting to unmout a
busy ATARAID member as if it was a file system. This is not a valid
thing to do, so disallow it.
Closes#75 - Errors with GPT on RAID 0 ATARAID array
This matches how other non-file systems are handled, by displaying the
access reference in the mount point column. For LVM Physical Volumes
the Volume Group name is displayed [1] and for an active Linux Software
RAID array the array device is displayed [2].
[1] 8083f11d84
Display LVM2 VGNAME as the PV's mount point (#160787)
[2] f6c2f00df7
Populate member mount point with SWRaid array device (#756829)
Closes#75 - Errors with GPT on RAID 0 ATARAID array
This stops GParted allowing overwrite operations (such as create
partition table or format with a whole device file system) being
performed on an ATARAID member while the array is actively using the
member.
Closes#75 - Errors with GPT on RAID 0 ATARAID array
The previous commit, made mdadm recognised IMSM and DDF type ATARAID
members get displayed as "linux-raid" (Linux Software RAID array
member). This was because of query method 1 in detect_filesystems().
Fix this now by exposing and using the fstype of the member from the
SWRaid_Info cache.
Closes#75 - Errors with GPT on RAID 0 ATARAID array
Since mdadm release 3.0 (2009-06-02) [1] it has also supported external
metadata formats IMSM (Intel Matrix Storage Manager) and DDF, previously
only managed by dmraid.
A number of distributions have switched to use mdadm and kernel MD
(Multiple Devices) driver for managing these Firmware / BIOS / ATARAID
arrays. These include: Fedora >= 14 [2], RHEL / CentOS >= 6 [3],
SLES >= 12 [4], Ubuntu >= 16.04 LTS.
Therefore additionally parse members in these ATARAID arrays included in
mdadm output, and when activated using the kernel MD driver, in file
/proc/mdstat. Add fstype to the SWRaid_Info cache records to
distinguish members apart. So far the rest of the GParted code
continues to treat all members as FS_LINUX_SWRAID. This will be
resolved in following commits.
Note that this in no way affects how GParted shows and partitions the
array device itself, even those managed by dmraid and use the GParted
DMRaid module. It only affects how GParted shows the member drives
themselves.
[1] mdadm ANNOUNCE-3.0 file
https://git.kernel.org/pub/scm/utils/mdadm/mdadm.git/tree/ANNOUNCE-3.0?h=mdadm-3.0
[2] Fedora 14, Storage Administration Guide, 12.5. Linux RAID Subsystem
https://docs.fedoraproject.org/en-US/Fedora/14/html/Storage_Administration_Guide/raid-subsys.html
"... Fedora 14 uses mdraid with external metadata to access ISW /
IMSM (Intel firmware RAID) sets. mdraid sets are configured and
controlled through the mdadm utility."
[3] RHEL 6, Storage Administration Guide, 17.3. Linux RAID Subsystem
https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html/storage_administration_guide/raid-subsys
"mdraid also supports other metadata formats, known as external
metadata. Red Hat Enterprise Linux 6 uses mdraid with external
metadata to access ISW / IMSM (Intel firmware RAID) sets. mdraid
sets are configured and controlled through the mdadm utility."
[4] SUSE Linux Enterprise Server 12 Release Notes, 7.2.3 Driver for IMSM
and DDF
https://www.suse.com/releasenotes/x86_64/SUSE-SLES/12/#fate-316007
"For IMSM and DDF RAIDs the mdadm driver is used unconditionally."
Closes#75 - Errors with GPT on RAID 0 ATARAID array
PATCHSET OVERVIEW
A user had a Firmware / BIOS / ATARAID array of 2 devices configured as
a RAID 0 (stripe) set. On top of that was a GPT with the OS partitions.
GParted displays the following errors on initial load and subsequent
refresh:
Libparted Error
(-) Invalid argument during seek for read on /dev/sda
[ Retry ] [ Cancel ] [ Ignore ]
Libparted Error
(-) The backup GPT table is corrupt, but the
primary appears OK, so that will be used.
[ Ok ] [ Cancel ]
This is an Intel Software RAID array which stores metadata at the end of
each member device, and so the first 128 KiB stripe of the set is stored
in the first 128 KiB of the first member device /dev/sda which includes
the GPT for the whole RAID 0 device. Hence when libparted reads member
device /dev/sda it finds a GPT describing a block device twice it's
size and in results the above errors when trying to read the backup GPT.
A more dangerous scenario occurs when using 2 devices configured in an
Intel Software RAID 1 (mirrored) set with GPT on top. On refresh
GParted display this error for both members, /dev/sda and /dev/sdb:
Libparted Warning
/!\ Not all of the space available to /dev/sda appears to be used,
you can fix the GPT to use all of the space (an extra 9554
blocks) or continue with the current setting?
[ Fix ] [ Ignore ]
Selecting [Fix] gets libparted to re-write the backup GPT to the end of
the member device, overwriting the ISW metadata! Do that twice and both
copies of the metadata are gone!
Worked example of this more dangerous mirrored set case. Initial setup:
# dmraid -s
*** Group superset isw_caffbiaegi
--> Subset
name : isw_caffbiaegi_MyMirror
size : 16768000
stride : 128
type : mirror
status : ok
subsets: 0
devs : 2
spares : 0
# dmraid -r
/dev/sda: isw, "isw_caffbiaegi", GROUP, ok, 16777214 sectors, data@ 0
/dev/sdb: isw, "isw_caffbiaegi", GROUP, ok, 16777214 sectors, data@ 0
# wipefs /dev/sda
offset type
---------------------------------------------
0x200 gpt [partition table]
0x1fffffc00 isw_raid_member [raid]
Run GParted and click [Fix] on /dev/sda. Now the first member has gone:
# dmraid -s
*** Group superset isw_caffbiaegi
--> *Inconsistent* Subset
name : isw_caffbiaegi_MyMirror
size : 16768000
stride : 128
type : mirror
status : inconsistent
subsets: 0
devs : 1
spares : 0
# dmraid -r
/dev/sdb: isw, "isw_caffbiaegi", GROUP, ok, 16777214 sectors, data@ 0
# wipefs /dev/sda
offset type
---------------------------------------------
0x200 gpt [partition table]
Click [Fix] on /dev/sdb. Now all members of the array are gone:
# dmraid -s
no raid disks
# dmraid -r
no raid disks
# wipefs /dev/sdb
offset type
---------------------------------------------
0x200 gpt [partition table]
So GParted must not run libparted partition table scanning on the member
devices in ATARAID arrays. Only on the array device itself.
In terms of the UI GParted must show disks which are ATARAID members as
whole disk devices with ATARAID member content and detect array busy
status to avoid allowing active members from being overwritten while in
use.
THIS COMMIT
Recognise ATARAID member devices and display in GParted as whole device
"ataraid" file systems. Because they are recognised as whole device
content ("ataraid" file systems) this alone stops GParted running the
libparted partition table scanning and avoids the above errors.
The list of dmraid supported formats is matched by the signatures
recognised by blkid:
$ dmraid -l
asr : Adaptec HostRAID ASR (0,1,10)
ddf1 : SNIA DDF1 (0,1,4,5,linear)
hpt37x : Highpoint HPT37X (S,0,1,10,01)
hpt45x : Highpoint HPT45X (S,0,1,10)
isw : Intel Software RAID (0,1,5,01)
jmicron : JMicron ATARAID (S,0,1)
lsi : LSI Logic MegaRAID (0,1,10)
nvidia : NVidia RAID (S,0,1,10,5)
pdc : Promise FastTrack (S,0,1,10)
sil : Silicon Image(tm) Medley(tm) (0,1,10)
via : VIA Software RAID (S,0,1,10)
dos : DOS partitions on SW RAIDs
$ fgrep -h _raid_member util-linux/libblkid/src/superblocks/*.c
.name = "adaptec_raid_member",
.name = "ddf_raid_member",
.name = "hpt45x_raid_member",
.name = "hpt37x_raid_member",
.name = "isw_raid_member",
.name = "jmicron_raid_member",
.name = "linux_raid_member",
.name = "lsi_mega_raid_member",
.name = "nvidia_raid_member",
.name = "promise_fasttrack_raid_member",
.name = "silicon_medley_raid_member",
.name = "via_raid_member",
As they are all types of Firmware / BIOS / ATARAID arrays, report all
members as a single "ataraid" file system type. (Except for
"linux_raid_member" in the above blkid source listing which is Linux
Software RAID).
Closes#75 - Errors with GPT on RAID 0 ATARAID array
The HACKING file should be hints for making changes to the code base and
associated processes. A overview of how GParted handled unallocated
space was not that. Also now the size of a JFS is accurately calculated
using JFS as an example of a file system with intrinsic unallocated
space is no longer valid. Therefore removed from the HACKING file.
Instead add the original commit message as an extended comment to method
calc_significant_unallocated_sectors().
Closes!50 - Calculate JFS size accurately
With the same minimum sized 16 MiB JFS used in the previous commit, now
mounted, GParted once again reports 1.20 MiB of unallocated space. This
is because the kernel JFS driver is also just reporting the size of the
Aggregate Disk Map (dmap) as the size of the file system [1].
Fix by reading the on disk JFS superblock to calculate the size of the
file system, but query the free space from the kernel using statvfs().
Need to query mounted JFS free space from the kernel because the on disk
dmap is not updated immediately so doesn't reflect recently used or
freed disk space.
For example, start with the 16 MiB JFS empty and mounted.
# echo -e 'dmap\nx\nquit' | jfs_debugfs /dev/sdb1 | fgrep dn_nfree
[2] dn_nfree: 0x00000000eaa [10] dn_agwidth: 1
# df -k /mnt/1
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sdb1 15152 136 15016 1% /mnt/1
Write 10 MiB of data to it:
# dd if=/dev/zero bs=1M count=10 of=/mnt/1/file_10M
10+0 records in
10+0 records out
1048760 bytes (10 MB, 10 MiB) copied, 0.0415676 s, 252 MB/s
Query the file system free space from the kernel and by reading the on
disk dmap figure:
# df -k /mnt/1
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sdb1 15152 10376 4776 69% /mnt/1
# echo -e 'dmap\nx\nquit' | jfs_debugfs /dev/sdb1 | fgrep dn_nfree
[2] dn_nfree: 0x00000000eaa [10] dn_agwidth: 1
# sync
# echo -e 'dmap\nx\nquit' | jfs_debugfs /dev/sdb1 | fgrep dn_nfree
[2] dn_nfree: 0x00000000eaa [10] dn_agwidth: 1
# umount /mnt/1
# echo -e 'dmap\nx\nquit' | jfs_debugfs /dev/sdb1 | fgrep dn_nfree
[2] dn_nfree: 0x000000004aa [10] dn_agwidth: 1
The kernel reports the updated usage straight away, but the on disk dmap
record doesn't get updated even by sync, only after unmounting.
This is the same fix as was previously done for EXT2/3/4 [2].
[1] Linux jfs_statfs() function
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/fs/jfs/super.c?h=v3.10#n142
[2] 3828019030
Read file system size for mounted ext2/3/4 from superblock (#683255)
Closes!50 - Calculate JFS size accurately
Upstream NILFS project calls the package nilfs-utils [1][2]. Arch Linux
/ CentOS / Fedora / OpenSUSE use the upstream name. However Debian /
Ubuntu name it nilfs-tools [3] instead.
Document the needed software as:
nilfs-utils / nilfs-tools
Upstream name first separated by slash from alternative names
distributions use.
[1] NILFS Download page
https://nilfs.sourceforge.io/en/download.html
[2] NILFS Public Git Repositories
https://nilfs.sourceforge.io/en/git_repos.html
[3] Debian package: nilfs-tools
https://packages.debian.org/sid/nilfs-tools
Running JFS read usage test on Fedora 30 fails like this:
$ ./test_SupportedFileSystems --gtest_filter='*ReadUsage/jfs'
...
[ RUN ] My/SupportedFileSystemsTest.CreateAndReadUsage/jfs
unknown file: Failure
C++ exception with description "std::bad_alloc" thrown in the test body.
[ FAILED ] My/SupportedFileSystemsTest.CreateAndReadUsage/jfs, where GetParam() = 17 (41833 ms)
However the same test passes on Fedora 29, Fedora 31 Beta, CentOS 7,
Debian 10 and Ubuntu 18.04 LTS.
Also running GParted on Fedora 30 crashes just the same when reading JFS
usage:
# gparted
GParted 1.0.0
configuration --enable-libparted-dmraid --enable-online-resize
libparted 3.2
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
/usr/bin/gparted: line 202: 19218 Aborted (core dumped) $BASE_CMD
Running jfs_debugfs to query the file system usage the same way GParted
does produces an infinite amount of repeating output:
# echo dm | jfs_debugfs /dev/sdb1
So jfs_debugfs gets stuck in an infinite loop inside the dmap subcommand
when it encounters EOF. GParted and the read JFS usage test read this
output until memory is exhausted and crash. This is exactly what was
happening in closed bug 794947. Even installed jfsutils from Fedora 29
on Fedora 30 and visa versa. jfs_debugfs still produced an infinite
amount of output on Fedora 30 and worked correctly on Fedora 29. So
it's not the build of jfsutils, but something in the OS that is making
the difference!
Anyway fix by providing the instruction to exit from the dmap
subcommand, and quit from jfs_debugfs itself, like this:
# echo -e 'dmap\nx\nquit' | jfs_debugfs /dev/sdb1
Bug 794947 - gparted hangs when sees JFS partition on discovering
partitions
Closes!49 - Add file system interface tests
GParted_Core::FILESYSTEMS and ::FILESYSTEM_MAP and the methods that
query and manipulate them are self-contained. Therefore move them into
a separate SupportedFileSystems module.
Also having a single class maintaining all FileSystem interface objects
will make testing all the file system types much easier as there will be
no need to duplicate this functionality in the test.
Closes!49 - Add file system interface tests
To match the renaming performed as part of Bug 741424 "Add support for
GPT partition names". In particular this is most closely similar to:
d480800600
Rename enum to OPERATION_LABEL_FILESYSTEM (#741424)
Separate udevsettle command was merged into udevadm back in udev 117
(released 2007-11-13) by:
225cb03bd8
udevadm: merge all udev tools into a single binary
The oldest supported distributions have these much newer versions of
udev:
Distro EOL udevadm -V
Debian 8 2020-Apr 215 (combined systemd and udev)
RHEL / CentOS 7 2024-Jun 219 (combined systemd and udev)
Ubuntu 16.04 LTS 2021-Apr 229 (combined systemd and udev)
Along with the previous patch "Avoid reading the same sector multiple
times in a row (!46)(#16)" internal detection of file system signatures
now reads the minimum number of sectors in increasing order.
As the code still considers the sector size, it now also reads less
sectors from devices with larger sectors to get all the signatures. So
on a 512 byte sector device it maximally seeks to and reads from these
byte offsets:
0, 512, 1024, 65536
and on a 4096 byte sector device, only from these byte offsets:
0, 65536
Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
Closes#16 - "invalid argument for seek()" error on very small (<=40KiB)
drives
GParted internal file system detection is reading the same sector
multiple times in a row. Add an optimisation to only read a sector if
it is different to the previously read sector.
Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
Closes#16 - "invalid argument for seek()" error on very small (<=40KiB)
drives
After the previous commit the lp_device structure pointer parameter is
only used to provide sector_size. Just pass that instead.
Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
Closes#16 - "invalid argument for seek()" error on very small (<=40KiB)
drives
GParted_Core::useable_device() doesn't change the pointed to PedDevice
structure. Pass it as const so the compiler enforces this.
Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
For every device that GParted scans, it determines if it can be used by
attempting to read the first sector. This is the sequence of events,
as reported in the previous two commit messages:
gparted |set_devices_thread(pdevices) pdevices->["/dev/sdb"]
...
gparted | useable_device(lp_device) lp_device->path="/dev/sdb"
gparted | ped_device_open()
libparted | open("/dev/sdb", O_RDWR) = 8
gparted | ped_device_read()
gparted | ped_device_close()
libparted | close(8)
udev(async)| KERNEL change /devices/.../sdb (block)
...
udev(async)| UDEV change /devices/.../sdb (block)
Note that these udev block device change events occur before the ones
waited for in the first commit "Wait for udev change on /dev/DISK when
querying whole device FS (!46)" so these were already waited for.
So because libparted only opens devices read-write this triggers a set
of udev change events for the whole block device, and if the device is
partitioned, also remove and re-add events for all the partitions.
Switch to using POSIX open(), read() and close() calls so read-only
access can be specified to avoid the unnecessary udev change events.
Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
After the previous fix, on my CentOS 7 VM, GParted is now often
reporting a warning that the /dev/PTN entry is missing when reading the
label of the file system in the first partition. This happens
regardless of the file system type.
Example error for EXT4:
e2label: No such file or directory while trying to open /dev/sdb1
Couldn't find valid file system superblock.
Example error for XFS:
/dev/sdb1: No such file or directory
fatal error -- couldn't initialize XFS library
As with the case in the previous commit, GParted gets the label via
blkid instead.
Again with debugging and sleeping in GParted, stracing the GParted
thread GParted_Core::set_devices_thread() and using 'udevadm monitor' to
watch udev events the following sequence of events is observed:
gparted |set_devices_thread(pdevices) pdevices->["/dev/sdb"]
gparted | ped_device_get("/dev/sdb")
libparted | open("/dev/sdb", O_RDONLY) = 8
libparted | close(8)
gparted | useable_device(lp_device) lp_device->path="/dev/sdb"
gparted | ped_device_open()
libparted | open("/dev/sdb", O_RDWR) = 8
gparted | ped_device_read()
gparted | ped_device_close()
libparted | close(8)
udev(async)| KERNEL remove /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL remove /devices/.../sdb/sdb2 (block)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| KERNEL add /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL add /devices/.../sdb/sdb2 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb1 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb2 (block)
udev(async)| UDEV change /devices/.../sdb (block)
udev(async)| UDEV add /devices/.../sdb/sdb1 (block)
udev(async)| UDEV add /devices/.../sdb/sdb2 (block)
gparted | set_device_from_disk(device, device_path="/dev/sdb")
gparted | get_device(device_path="/dev/sdb", ..., flush=true)
gparted | ped_device_get()
gparted | flush_device(lp_device) lp_device->path="/dev/sdb"
gparted | ped_device_open()
libparted | open("/dev/sdb", O_RDWR) = 8
gparted | ped_device_sync()
gparted | ped_device_close()
libparted | close(8)
udev(async)| KERNEL remove /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL remove /devices/.../sdb/sdb2 (block)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| KERNEL add /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL add /devices/.../sdb/sdb2 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb1 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb2 (block)
udev(async)| UDEV change /devices/.../sdb (block)
udev(async)| UDEV add /devices/.../sdb/sdb1 (block)
udev(async)| UDEV add /devices/.../sdb/sdb2 (block)
gparted | settle_device()
gparted | Utils::execute_command("udevadm settle --timeout=1")
gparted | detect_filesystem(lp_device, NULL, ...) lp_device->path="/dev/sdb"
gparted | detect_filesystem_internal(lp_device, NULL) lp_device->path="/dev/sdb"
gparted | ped_device_open()
libparted | open("/dev/sdb", O_RDWR) = 8
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_read()
gparted | ped_device_close()
libparted | close(8)
udev(async)| KERNEL remove /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL remove /devices/.../sdb/sdb2 (block)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| KERNEL add /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL add /devices/.../sdb/sdb2 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb1 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb2 (block)
udev(async)| UDEV change /devices/.../sdb (block)
udev(async)| UDEV add /devices/.../sdb/sdb1 (block)
udev(async)| UDEV add /devices/.../sdb/sdb2 (block)
gparted | get_disk(lp_device, lp_disk, strict=false)
gparted | ped_disk_new(lp_device) lp_device->path="/dev/sdb"
libparted | open("/dev/sdb", O_RDWR) = 8
libparted | close(8)
udev(async)| KERNEL remove /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL remove /devices/.../sdb/sdb2 (block)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| KERNEL add /devices/.../sdb/sdb1 (block)
udev(async)| KERNEL add /devices/.../sdb/sdb2 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb1 (block)
udev(async)| UDEV remove /devices/.../sdb/sdb2 (block)
udev(async)| UDEV change /devices/.../sdb (block)
udev(async)| UDEV add /devices/.../sdb/sdb1 (block)
udev(async)| UDEV add /devices/.../sdb/sdb2 (block)
gparted | set_device_partitions()
gparted | detect_filesystem()
gparted | set_partition_label_and_uuid(partition) partition.get_path()="/dev/sdb1"
gparted | read_label()
gparted | ext2::read_label()
gparted | Utils::execute_command("e2label /dev/sdb1")
So in this case for a partitioned drive, libparted ped_disk_new() is
opening and closing the device read-write and triggering the udev remove
and add partition block special entries exactly when the label is trying
to be read from the first partition, causing the device missing error.
The call chain is:
set_devices_thread()
set_device_from_disk()
get_disk()
ped_disk_new()
Looking at the source for ped_disk_new() [1] it is calling
ped_device_open() and ped_device_close(), hence why it is opening the
device read-write and triggering the udev events.
Looking back at commit "Wait for udev to recreate /dev/PTN entries when
calibrating (#762941)" [2] and re-testing it, the udev events were also
caused by ped_disk_new() with this call chain:
apply_operation_to_disk()
calibrate_partition()
get_disk()
ped_disk_new()
Fix this probe case and keep the fix for the previous calibrate case by
moving the waiting for udev events from calibrate_partition() into
get_disk(), right after ped_disk_new(). The maximum wait time is
shortened to the shorter 1 second probing timeout to avoid the longer
10 second apply timeout in this probing case. The calibrate case commit
message said "The longest I have seen udev take to do this is 0.80
seconds in a VM".
[1] parted/libparted/disk.c:ped_disk_new()
http://git.savannah.gnu.org/cgit/parted.git/tree/libparted/disk.c?id=v3.2#n169
[2] fd9013d5f6
Wait for udev to recreate /dev/PTN entries when calibrating
(#762941)
Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
GParted nearly always shows this warning against a whole disk device
FAT32 file system on my development VMs:
plain floppy: device "/dev/sdb" busy (Resource temporarily
unavailable): Cannot initialize '::'
mlabel: Cannot initialize drive
However GParted does get the label via blkid instead. Occasionally
everything works correctly and there is no warning.
Never found a similar error for any other file system on a whole disk
device. The timing never seems to clash.
Remember that when a writable file descriptor of a disk device is
closed, udev events are triggered which update the kernel and /dev
entries for any partition changes. (This is in addition to coding which
has always existed in the partitioning tools, including fdisk and
parted, to inform the kernel, but not create /dev entries, of partition
changes).
With debugging and sleeping in GParted, stracing the GParted thread
GParted_Core::set_devices_thread() and using 'udevadm monitor' to watch
udev events the following sequence of events is observed:
gparted |set_devices_thread(pdevices) pdevices->["/dev/sdb"]
gparted | ped_device_get("/dev/sdb")
libparted | open("/dev/sdb", O_RDONLY) = 8
libparted | close(8)
gparted | useable_device(lp_device) lp_device->path="/dev/sdb"
gparted | ped_device_open()
libparted | open("/dev/sdb", O_RDWR) = 8
gparted | ped_device_read()
gparted | ped_device_close()
libparted | close(8)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| UDEV change /devices/.../sdb (block)
udev(async)| UDEV change /devices/.../sdb (block)
gparted | set_device_from_disk(device, device_path="/dev/sdb")
gparted | get_device(device_path="/dev/sdb", ..., flush=true)
gparted | ped_device_get()
gparted | flush_device(lp_device) lp_device->path="/dev/sdb"
gparted | ped_device_open()
libparted | open("/dev/sdb", O_RDWR) = 8
gparted | ped_device_sync()
gparted | ped_device_close()
libparted | close(8)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| KERNEL change /devices/.../sdb (block)
udev(async)| UDEV change /devices/.../sdb (block)
udev(async)| UDEV change /devices/.../sdb (block)
gparted | set_device_one_partition()
gparted | set_partition_label_and_uuid()
gparted | read_label()
gparted | fat16::read_label()
gparted | Utils::execute_command("mlabel -s :: -i /dev/sdb")
This sequence of events only shows which close()s by libparted trigger
udev events. It does not show that processing those events are
asynchronous and overlap with GParted executing mlabel, causing the
device busy error.
As libparted's ped_device_open() doesn't offer a way to specify that a
device will only be used for reading, it always opens it read-write
[1][2]. Hence this sequence in GParted_Core::flush_device():
ped_device_open()
ped_device_sync()
ped_device_close()
always triggers udev change events on the whole disk device. Fix by
waiting for those udev events to complete.
[1] libparted documentation, PedDevice, ped_device_open()
https://www.gnu.org/software/parted/api/group__PedDevice.html#ga7c87bd06e76553c8ff3262b5e6ca256
[2] parted/libparted/arch/linux.c:linux_open()
http://git.savannah.gnu.org/cgit/parted.git/tree/libparted/arch/linux.c?id=v3.2#n1628Closes!46 - Whole device FAT32 file system reports device busy warning
from mlabel
The last distribution to include the separate 'udevinfo' command was
RHEL / CentOS 5 with udev 095. All supported distributions have much
newer versions of udev and instead have the 'udevadm' command.
A user reported that GParted was slow to refresh on FAT32 file systems:
"can take very long, up to several minutes; can be reproduced by running
dosfsck manually". This is because the file system was large and almost
full, and GParted performs a file system check just to to report the
file system usage.
Created a 4 GiB FAT32 file system and almost filled it with 4 KiB files,
just over 970,000 files.
# df -k /mnt/2
Filesystem 1K-blocks Used Available Used% Mounted on
/dev/sdb2 4186108 39155384 270724 94% /mnt/2
# df -i /mnt/2
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/sdb2 0 0 0 - /mnt/2
# find /mnt/2 -type f -print | wc -l
971059
# find /mnt/2 -type d -print | wc -l
1949
Testing performance of the current fsck.fat:
# time fsck.fat -n -v /dev/sdb2 | \
> egrep 'bytes per logical sector|bytes per cluster|sectors total|clusters$'
512 bytes per logical sector
4096 bytes per cluster
8388608 sectors total
/dev/sdb2: 973008 files, 978846/1046527 clusters
real 0m11.552s
user 0m2.183s
sys 0m7.547s
Free sectors in the file system according to fsck.fat:
(1046527 - 978846) * 4096 / 512 = 541448 sectors
Repeating this test while also using 'blktrace /dev/sdb2' and Ctrl-C
around the test in a separate terminal, reports these numbers of I/Os
being performed:
Read requests Read bytes
15,563 165 MiB
Prior to this commit [1] from 0.0.9, GParted used libparted's
ped_file_system_get_resize_constraint() to report the minimum size to
which a FAT* file system can be resized. Use this test program [2] to
performance test this method:
# time ./fscons /dev/sdb2
dev=/dev/sdb2
sector_size=512
min_size=7909522
max_size=8388608
real 0m2.673s
user 0m0.070s
sys 0m1.834s
Free sectors in the file system according to libparted
ped_file_system_get_resize_constraint():
8388608 - 7909522 = 479086 sectors
blktrace reports these numbers of I/Os being performed:
Read requests Read bytes
7,821 71 MiB
So using libparted resize constraint is a bit faster but is still
reading too much data and is really too slow. Also when testing GParted
using this libparted method against a corrupted FAT32 file system, on
every refresh, one popup dialog is displayed for each error libparted
detects with the file system, each of which needs acknowledgement.
Example popup:
Libparted Error
\DIRNAME\FILENAME.EXT is 107724k, but is has 1920
clusters (122880k).
[ Cancel ][ Ignore ]
There could be a huge number of such errors in a corrupted file system.
Not really suitable for use by GParted.
Test the performance of mtools' minfo command to report the file system
figures:
# time minfo -i /dev/sdb2 :: | \
> egrep 'sector size:|cluster size:|small size:|big size:|free clusters='
sector size: 512 bytes
cluster size: 8 sectors
small size: 0 sectors
big size: 8388608 sectors
free clusters=67681
real 0m0.013s
user 0m0.004s
sys 0m0.019s
Free sectors in the file system according to minfo:
67681 * 8 = 541448 sectors
blktrace reports these numbers of I/Os being performed by minfo:
Read requests Read bytes
1 16 KiB
This matches with minfo just reading information from the BPB (BIOS
Parameter Block) [3] from sector 0 and the FS Information Sector [4]
usually in sector 1. Note that the free cluster figure reported by
minfo comes from the FS Information Sector because it only reports it
for FAT32 file systems, not for FAT16 file systems. Scanning the File
Allocation Table (FAT) [5] to count free clusters is exactly what mdir,
without the '-f' (fast) flag, does. Test the performance of mdir:
# export MTOOLS_SKIP_CHECK=1
# time mdir -i /dev/sdb2 ::/ | fgrep 'bytes free'
277 221 376 bytes free
real 0m0.023s
user 0m0.011s
sys 0m0.023s
Free sectors in the file system according to mdir:
277221376 / 512 = 541448 sectors
blktrace reports these number of I/Os being performed by mdir:
Read requests Read bytes
5 448 KiB
So minfo and mdir together provide the needed information and are 2 to 3
orders of magnitude faster because they only read the needed BPB and FAT
data from the drive. Use these together to read the file system usage.
[1] 61cd0ce778
lots of stuff and cleanups, including fixing getting used/unused
space of hfs/hfs+/fat16/fat32
[2] fscons.c
/* FILE: fscons.c
* SYNOPSIS: Report libparted's FS resize limits.
* BUILD: gcc -o fscons fscons.c -lparted -lparted-fs-resize
*/
int main(int argc, const char *argv[])
{
PedDevice* dev = NULL;
PedDisk* tab = NULL;
PedPartition* ptn = NULL;
PedFileSystem* fs = NULL;
PedConstraint* cons = NULL;
if (argc != 2)
{
fprintf(stderr, "Usage: fscons BLOCKDEV\n");
exit(1);
}
dev = ped_device_get(argv[1]);
if (dev == NULL)
{
fprintf(stderr, "ped_device_get(\"%s\") failed\n", argv[1]);
exit(1);
}
printf("dev=%s\n", dev->path);
printf("sector_size=%ld\n", dev->sector_size);
tab = ped_disk_new(dev);
if (tab == NULL)
{
fprintf(stderr, "ped_disk_new(dev) failed\n");
exit(1);
}
ptn = ped_disk_get_partition_by_sector(tab, 0);
if (ptn == NULL)
{
fprintf(stderr, "ped_disk_get_partition(tab, 0) failed\n");
exit(1);
}
fs = ped_file_system_open(&ptn->geom);
if (fs == NULL)
{
fprintf(stderr, "ped_file_system_open(&ptn->geom) failed\n");
exit(1);
}
cons = ped_file_system_get_resize_constraint(fs);
if (cons == NULL)
{
fprintf(stderr, "ped_file_system_get_resize_constraint(fs) failed\n");
exit(1);
}
printf("min_size=%ld\n", cons->min_size);
printf("max_size=%ld\n", cons->max_size);
ped_constraint_destroy(cons);
ped_file_system_close(fs);
ped_disk_destroy(tab);
ped_device_destroy(dev);
return 0;
}
[3] Design of the FAT file system, BIOS Parameter Block
https://en.wikipedia.org/wiki/Design_of_the_FAT_file_system#BIOS_Parameter_Block
[4] Design of the FAT file system, FS Information Sector
https://en.wikipedia.org/wiki/Design_of_the_FAT_file_system#FS_Information_Sector
[5] Design of the FAT file system, File Allocation Table
https://en.wikipedia.org/wiki/Design_of_the_FAT_file_system#File_Allocation_Table
Bug 569921 - dosfsck -n delays device scan
When GParted is configured with `--disable-doc` the help documentation
is neither built, nor installed.
With this configuration the Help -> Contents menu displays a message
dialog indicating the "Documentation is not available".
On GNOME 3 no title is shown; however, on some other graphic toolkits
/ window managers an unspecified title is shown.
For example on GParted Live with Fluxbox the following is displayed:
Unnamed
-------
Documentation is not available
This build of gparted is configured without documentation.
Documentation is available at the project web site.
https://gparted.org
OK
To address the unspecified title on non-GNOME 3 graphic toolkits, add
a title that works with and without GNOME 3.
Closes!45 - Add missing window title to Help Contents dialog
All the dialogs which compose new or modified partition boundaries
create those partitions within the boundaries of the device. Therefore
trimming the partition boundaries to device boundaries never happens.
So replace this trimming with bug error reports.
Also add bug prefixes to the other error messages in
GParted_Core::valid_partition() too.
Closes#48 - Error when moving locked LUKS-encrypted partition
The previous commit moved the code from GParted_Core to
Dialog_Base_Partition class without making a single formatting change to
ensure the code was guaranteed to work the same within that larger
commit. Now reformat the code to current layout standards. Making it a
separate commit simplifies the effort for both changes and improves
bisectability.
Additionally to be sure there were no code changes,
Dialog_Base_Partition.cc was compiled to assembler code with and without
this change applied and the resultant assembler code compared. There
were no differences in the generated assembler code.
Start with the make generated g++ command for compiling
Dialog_Base_Partition.o; (1) remove the '-g' flag as inserted debugging
directives do differ; and (2) replace '-c -o Dialog_Base_Partition.o'
with '-S -o Dialog_Base_Partition.s' to produce assembler output instead
of object code.
Closes#48 - Error when moving locked LUKS-encrypted partition
Move snap_to_*() method calls from the point when all operations are
added to the list, to earlier when Resize/Move, Paste (into new) and
Create New dialogs are composing the new partition objects. In
particular for the Resize/Move operation, to just before updating the
file system usage.
This change finally resolves this bug.
Because of the dialog call chains into Dialog_Base_Partition,
snap_to_alignment() must be added into:
* Dialog_Base_Partition::prepare_new_partition() for the Resize/Move and
Paste (into new) dialogs; and
* Dialog_Partition_New::Get_New_Partition() for the Create New dialog.
Closes#48 - Error when moving locked LUKS-encrypted partition
The current device has to be passed to the dialog constructors and then
on to the Dialog_Base_Constructor. Note that the Dialog_Partition_New
constructor is already passed the current device, however it still needs
to pass it on to Dialog_Base_Constructor.
This is ready so that snap_to_*() methods can access the current device
when they are called from within these dialogs.
* Pass Parameter to Base Class Constructor while creating Derived class
Object
https://stackoverflow.com/questions/16585856/pass-parameter-to-base-class-constructor-while-creating-derived-class-objectCloses#48 - Error when moving locked LUKS-encrypted partition
Move setting of the new_partition object file system usage to after
everything else, specifically after free_space_before and strict_start.
This is because snap_to_*() use free_space_before and strict_start and
snap_to_alignment() is going to be called before the file system usage
is updated to avoid the error in this bug report.
Closes#48 - Error when moving locked LUKS-encrypted partition
None of the snap_to_*() methods report any errors so remove the unused
error parameter and return value.
Closes#48 - Error when moving locked LUKS-encrypted partition
PATCHSET OVERVIEW
A user had 2 adjacent partitions which were aligned to multiples of
33553920 bytes (32 MiB - 512 bytes), not MiB or cylinders. As far as
GParted is concerned this is not aligned. The second partition
contained closed LUKS encrypted data. Recreate this setup with:
# truncate -s 200G /tmp/disk.img
# losetup -f --show /tmp/disk.img
/dev/loop0
# sfdisk -u S /dev/loop0 << EOF
65535 2162655 83
2228190 78904140 83
EOF
# partprobe /dev/loop0
# echo -n badpassword | cryptsetup luksFormat /dev/loop0p2 -
When trying to move the second LUKS encrypted partition to the right by
any amount, with the default MiB alignment, GParted displays this error
dialog and fails to even queue the operation:
Could not add this operation to the list
A partition with used sectors (78907392) greater than its
length (78905344) is not valid
[ OK ]
Overview of the steps involved:
1. The Resize/Move dialog composed a new partition to start a whole
multiple of MiB after the end of the previous non-aligned partition.
The new partition also had it's size increased to a whole multiple of
MiB, to 78907392 sectors (38529 MiB) which was 1.59 MiB larger than
before. Neither the start or end of the new partition are aligned at
this point.
2. Win_GParted::activate_resize() applied the change back to the closed
LUKS partition object, additionally making the used sectors equal to
the partition size.
(To match the fact that when opened the LUKS mapping it will
automatically fill the new larger partition size).
3. GParted_Core::snap_to_mebibyte() then aligned the partition start and
end to whole MiB boundaries, reducing the partition size in the
process to 78905344 (38528 MiB).
4. GParted_Core::snap_to_alignment() reported the error saying that it
couldn't add the operation to the list because it was invalid to have
the file system used sectors larger than the partition size.
Fix this by having the snap to alignment adjustments applied before the
dialogs update any associated file system usage. Specifically the
Resize/Move, Paste (into new) and Create New dialogs as these are the
only ones which either create or modify partition boundaries.
Validation done by snap_to_alignment() will continue to occur at the
current point when the operation is added to the list.
THIS COMMIT
snap_to_alignment() is doing two different jobs, it is (1) optionally
adjusting the new partition boundaries for MiB or Cylinder alignment;
and (2) checking that the partition boundaries and file system usage are
valid.
Split those into two different functions (1) snap_to_alignment() and
(2) valid_partition(). For now valid_partition() still calls
snap_to_alignment() so there is no functional change with this commit.
Closes#48 - Error when moving locked LUKS-encrypted partition
After an NTFS file system has been resized the command GParted currently
uses to read the file system usage fails like this:
# ntfsinfo --mft /dev/sdb1
Volume is scheduled for check.
Please boot into Windows TWICE, or use the 'force' option.
NOTE: If you had not scheduled check and last time accessed this volume
using ntfsmount and shutdown system properly, then init scripts in your
distribution are broken. Please report to your distribution developers
(NOT to us!) that init scripts kill ntfsmount or mount.ntfs-fuse during
shutdown instead of proper umount.
Failed to open '/dev/sdb1'.
Fix by added the '--force' flag as described in the error message.
Closes#57 - NTFS Resize results in Partition Information Warning on
Refresh
GParted uses ntfsinfo to read the NTFS file system usage. However when
ntfsinfo fails with a non-zero exit status GParted reports stdout twice,
rather than stdout and stderr. Correct this.
Closes#57 - NTFS Resize results in Partition Information Warning on
Refresh
With the same case as from the previous commit, the very long "Mounted
on ..." text is now wrapped, but the text may not be left justified.
Slowly adjust the dialog width and see how the text wrapping is updated
to fit the size adjustment but the text is centred rather than left
justified.
This is because setting the halign property to Gtk::ALIGN_START does not
guarantee left alignment of text for wrapped or ellipsized Gtk::Labels.
Use the xalign property instead.
To set the xalign property there is a method in the GtkMisc (Gtk::Misc)
base class:
gtk_misc_set_alignment (Gtk::Misc::set_alignment)
However, GtkMisc (Gtk::Misc) was deprecated in Gtk 3.14 (Gtkmm 3.14)
and in Gtk 3.16 (gtkmm 3.16) set_alignment() was replaced with the
introduction of two new methods:
gtk_label_set_xalign (Gtk::Label::set_xalign)
gtk_label_set_yalign (Gtk::Label::set_yalign)
Add a check for Gtkmm method Gtk::Label::set_xalign() in configure.ac
and use it when available.
References:
[1] Gtk3 Reference Documentation - gtk_misc_set_alignment()
https://developer.gnome.org/gtk3/stable/GtkMisc.html#gtk-misc-set-alignment
"gtk_misc_set_alignment has been deprecated since version 3.14 and
should not be used in newly-written code. Use GtkWidget's alignment
("halign" and "valign") and margin properties or GtkLabel's
"xalign" and "yalign" properties."
[2] Gtkmm 3.16 Gtk::Misc Class Reference, set_alignment() method
https://developer.gnome.org/gtkmm/3.16/classGtk_1_1Misc.html#a52b2675874cf46a3097938756b9fe9e8
[3] GNOME BugZilla - EmptyBoxes: instructions_label's alignment is off
https://bugzilla.gnome.org/show_bug.cgi?id=735841
[4] Gtk commit from 2014-09-16:
GtkLabel: add x/yalign properties
https://gitlab.gnome.org/GNOME/gtk/commit/d39424fc
[5] Gtk3 Reference Documentation - gtk_label_set_xalign()
https://developer.gnome.org/gtk3/stable/GtkLabel.html#gtk-label-set-xalign
[6] Gtkmm 3.16 Gtk::Label Class Reference, set_xalign() method
https://developer.gnome.org/gtkmm/3.16/classGtk_1_1Label.html#acee7d4e87d7cc14080a7b8ded5f84e5eCloses!40 - Limit wrapping labels
Opening the Partition Information dialog for a file system mounted on a
very long mount point, or on openSUSE which mounts the OS from 10 btrfs
subvolumes from the same partition, will cause the dialog to be very
wide as the "Mounted on ..." text is not wrapped.
Back in Gtk2, when width_chars / max_width_chars were not set, wrapping
labels had a default width beyond which text wrapped onto a new line
[1].
For Gtk3 this default width was first reworked a bit [2], and then was
removed for the very early Gtk3 3.0.10 release [3].
It is recommended that applications explicitly set default values,
otherwise wrapping labels never wrap when requesting their natural
allocation.
References:
[1] Gtk 2.24.32 source code - gtk/gtklabel.c:2975
https://gitlab.gnome.org/GNOME/gtk/blob/2.24.32/gtk/gtklabel.c#L2975
"This long string gives a good enough length for any line to
have."
[2] Gtk commit from 2010-04-21:
680d7762ba
Make sure not to base the minimum size on "max-width-chars", only
the natural size.
"This string is just about long enough."
[3] Gtk commit from 2011-04-17:
c8ce1106c1
label: Don't try to guess a label's size
People should use window default sizes or label
width-chars/max-width-chars to find the ideal layout for a label
instead of relying on magic.
Closes!40 - Limit wrapping labels
Before Gtk 3.22 GtkScrolledWindow propagated natural size to its
Children and so on to descendants. In Gtk 3.22 this was changed to
always request the minimum size. This was done because it is believed
to be a safer default (gives a better behaviour) in case of dynamic
content inside the scrolled window, that is, content that may change
allocated size. [1][2][3]
When the scrolled window content is not dynamic the natural size is
preferable because it gives a better looking layout and without any
downside.
In the case of GParted content which is not dynamic, so request the
scrolled windows to allocate children at natural sizes for Gtk >= 3.22.
The benefits of natural size allocation are evident in presence of
wrapping labels (for example inside the "Partition Info" dialog), that
with the minimum size request likely end up taking a very small width.
References:
[1] Gtk commit from 2016-08-31:
GtkScrolledWindow: Make propagation of natural child sizes optional
0984d1622d
"Making propagation of child natural sizes mandatory (or default,
even) was evidently a mistake as this causes dynamic content in a
scrolled window to resize it's parent when the scrolled window is
competing for space with an adjacent widget."
[2] Gtk 3.22 Reference Documentation -
gtk_scrolled_window_set_propagate_natural_width
https://developer.gnome.org/gtk3/3.22/GtkScrolledWindow.html#gtk-scrolled-window-set-propagate-natural-width
[3] Gtkmm 3.24 Gtk::ScrolledWindow Class Reference,
set_propagate_natural_width() method
https://developer.gnome.org/gtkmm/3.24/classGtk_1_1ScrolledWindow.html#a2d4cb945688ecb8739efd70b18742779
[4] Gtkmm 3.21.6 NEWS
https://gitlab.gnome.org/GNOME/gtkmm/blob/3.21.6/NEWS
"ScrolledWindow: Added get/set_propagate_natural_height/width() and
the properties."
Closes!39 - Always request natural size inside Gtk::ScrolledWindow
Gtk::Table was deprecated in Gtk 3.4.0. Replace with Gtk::Grid.
This commit makes the change for Dialog_Partition_Info.
Closes!25 - Modern Gtk3 - part 1
Gtk::Table was deprecated in Gtk 3.4.0. Replace with Gtk::Grid.
This commit makes the change for Dialog_Base_Partition.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_Progress.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_Rescue_Data.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for DialogPasswordEntry.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_Partition_Name.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_FileSystem_Label.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for DialogFeatures.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_DiskLabel.cc.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_Partition_Info.{h,cc}.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for Dialog_Base_Partition.{h,cc}.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2. Replace with
plain Gtk::Box.
This commit makes the change for HBoxOperations.{h,cc}.
Closes!25 - Modern Gtk3 - part 1
Gtk::HBox and Gtk::VBox were deprecated in Gtkmm 3.2 [1]. Replace with
plain Gtk::Box.
This commit makes the change for Win_GParted.{h,cc}.
[1] Gtkmm 3.2.0 NEWS file (actually first included in gtkmm 3.1.6
unstable)
https://gitlab.gnome.org/GNOME/gtkmm/blob/3.2.0/NEWS#L91
Gtk:
* All H* or V* specialized classes have been deprecated, to
match the deprecations in the GTK+ C API. You should now
set the orientation instead.
This includes HBox, VBox, HButtonBox, VButtonBox, HPaned,
VPaned, HScale, VScale, HSeparator, VSeparator, HScrollbar and
VScrollbar.
Closes!25 - Modern Gtk3 - part 1
The Gdk::RGBA data type was introduced to replace Gdk::Color in
Gtkmm 3.0 [1], with Gdk::Color being deprecated in Gtkmm 3.10 [2].
With this commit we make the change to Gdk::RGBA data type which is the
modern replacement to Gdk::Color. Gdk::RGBA can be used almost as a
drop in replacement because it keeps most of the Gdk::Color interface.
Also, this commit removes the C Gtk call introduced during the
port-to-gtk3 patchset by commit:
5379352766
repare-for-gtk3: Prepare for removal of Gtk::Widget::modify_fg() (#7)
[1] Gtkmm 3.0.1 NEWS file
https://gitlab.gnome.org/GNOME/gtkmm/blob/3.0.1/NEWS#L48
* RGBA replaces Color, though Color still exists because it is
used by TextView. We hope to deprecated Color completely in
gtkmm 3.2.
[2] Gtkmm 3.10.0 NEWS file
https://gitlab.gnome.org/GNOME/gtkmm/blob/3.10.0/NEWS#L127
Gdk:
* Deprecate Color.
Closes!25 - Modern Gtk3 - part 1
Mostly add, but also remove, #includes so both DialogFeatures.h and .cc
include exactly the header files each needs to get the definitions they
use.
Header file #include guards are there to specifically enable this.
To better reflect that it is loading the supported actions for one file
system into the treeview, just like it's parent load_filesystems() is
initiating the loading for all the file systems.
Select a partition and look at the available actions in the Partition
menu. Then add or remove some commands which that particular file
system uses and rescan to detect those changes. Open the Partition menu
again. It doesn't reflect the changes of supported actions seen in the
File System Support dialog. Select a different partition and then
select the original partition again. Now the available actions in the
Partition menu reflect the changes of supported actions.
Have been testing by adding and removing /sbin/e2label to add and
remove EXT2/3/4 file system labelling support just because that feature
has existed for a very long time and EXT2/3/4 are displayed near the top
of the File System Support dialog. Tested this minor issue existed as
far back as GParted 0.3.7.
Fix by simply also refreshing the valid operations to update the
Partition menu after updating the found file system specific commands.
Closes!38 - Fixes for minor issues with File System Support rescanning
Open the File System Support dialog, either add or remove some file
system specific commands used by GParted and press the
[Rescan For Supported Actions] button. The supported actions don't
change. However after just closing and reopening the dialog, the
supported actions do reflect the added or removed file system specific
commands.
Bisected to this commit:
4d6d464664
Display "other" in the File System Support dialog (!13)
The problem is that commit made a subset copy of the
GParted_Core::FILESYSTEMS vector, obtained from get_filesystems(), so
when the rescan ran and the FILESYSTEMS vector was updated with new
supported actions, the dialog still displayed the original subset copy,
so didn't reflect the changed supported actions.
Fix by passing a reference to the GParted_Core::FILESYSTEMS vector,
obtained from get_filesystems(), and perform the necessary filtering
inside the dialog, like before the above faulty commit. Additionally
finding and adding "other" file system to the end of the list.
Closes!38 - Fixes for minor issues with File System Support rescanning
The oldest supported distributions have these versions of ntfs-3g /
ntfsprogs.
Distro EOL ntfs-3g / ntfsprogs
- Debian 8 2020-Jun 2014.2.16AR.2
- RHEL / CentOS 7 2024-Jun 2017.3.23
- Ubuntu 14.04 LTS 2019-Apr 2013.1.13AR.1
The oldest version of ntfs-3g / ntfsprogs on Ubuntu 14.04 LTS includes
support for the --new-serial option.
$ ntfslabel -V
ntfslabel v2013.1.13AR.1 (libntfs-3g) - Display, or set, the label for an NTFS Volume.
$ ntfslabel --help | grep -- --new-serial
--new-serial Set a new serial number
Therefore it is no longer necessary to check for this option as it is
always available. The worst case scenario is that some how an old
version of ntfslabel is used which doesn't support this option. In such
a case GParted goes from not supporting changing the UUID to claiming
support, but presumably it would fail with an error reporting unknown
option when applied. Arguably better from an end user support point of
view.
A user reported GParted was slow to refresh and timing ntfsresize to
query his file systems found that it was taking 4.7 seconds and 9.2
seconds for sizes 31.7 GiB and 49 GiB NTFS file systems respectively.
Almost 14 seconds just to read the usage.
Created a 4 GiB NTFS and filled it with as many 4 KiB files as possible,
just over 800,000 files.
# df -k /mnt/2
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sdb2 4194300 4193860 440 100% /mnt/2
# df -i /mnt/2
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/sdb2 819640 808591 11049 99% /mnt/2
Testing perform of ntfsresize:
# time ntfsresize --info --force --no-progress-bar /dev/sdb2 | \
> egrep 'Current volume size|resize at|Cluster size'
Cluster size : 4096 bytes
Current volume size: 4294963712 bytes (4295 MB)
You might resize at 4294516736 bytes (freeing 450560 bytes).
real 0m5.231s
user 0m2.072s
sys 0m3.014s
Computation of figures:
Clusters per volume = 4294963712 / 4096 = 1048575.125
Free clusters = (4294963712 - 4294516736) / 4096 = 109.125
Testing performance of ntfscluster, as used before this commit [1] from
GParted 0.3 in 2006:
# time ntfscluster --force /dev/sdb2 | \
> egrep 'bytes per cluster|bytes per volume|clusters per volume|clusters of free space'
...
bytes per cluster : 4096
bytes per volume : 4294963200
clusters per volume : 131071
clusters of free space : 110
real 0m4.243s
user 0m1.629s
sys 0m2.587s
Note that the clusters per volume figure reported by ntfscluster is
wrong. 4294963200 / 4096 = 1048575, not 131071.
Testing performance using ntfsinfo:
# time ntfsinfo --mft /dev/sdb2 | \
egrep 'Cluster Size|Volume Size in Clusters|Free Clusters'
Cluster Size: 4096
Volume Size in Clusters: 1048575
Free Clusters: 110 (0.0%)
real 0m0.022s
user 0m0.012s
sys 0m0.018s
Repeating the above tests while also using 'btrace /dev/sdb2' and Ctrl-C
around each test via a separate terminal, reports these numbers of I/Os
being performed:
Command Read requests Read bytes
- ntfsresize 2,695 1116 MiB
- ntfscluster 2,685 1116 MiB
- ntfsinfo 13 2208 KiB
No wonder that ntfsresize and ntfscluster take a long time, they read
just over 1 GiB of data from the 4 GiB file system, where as ntfsinfo
only reads just over 2 MiB.
Switch to using ntfsinfo to report file system usage.
[1] 9d956594d6
replaced ntfscluster with ntfsresize (see #350789)
Closes#47 - GParted is slow refreshing NTFS file systems
With that same commit in parted 1.9 [1], libparted only recognised these
linux-swap names via deprecated aliases:
linux-swap(old)
linux-swap(new)
but does accept this name as a current alias:
linux-swap
for:
linux-swap(v1)
Demonstration:
# parted -v
parted (GNU parted) 2.1
...
# parted /dev/sdc mkfs yes 1 "linux-swap(new)" unit s print
...
[0] filesys.c:148 (ped_file_system_type_get(): File system alias linux-swap(new) is deprecated
...
Number Start End Size Type File system Flags
1 2048s 2099199s 2097152s primary linux-swap(v1)
# parted /dev/sdc mkfs yes 1 "linux-swap" unit s print
...
Number Start End Size Type File system Flags
1 2048s 2099199s 2097152s primary linux-swap(v1)
Again as GParted now requires libparted 2.2 or later:
1) Stop using alternative "linux-swap(new)" name as that is deprecated
by libparted.
2) Also stop using alternative "linux-swap(v1)" name as that code is
never used because libparted recognised the GParted "linux-swap"
name as a current alias.
[1] http://git.savannah.gnu.org/cgit/parted.git/commit/?id=cfafa4394998a11f871a0f8d172b13314f9062c2
Rationalise linux-swap fs names, and add a "linux-swap" alias
With this commit in parted 1.8.3 [1], libparted changed from reporting
the name of Linux swap as:
linux-swap
to reporting either:
linux-swap(old)
linux-swap(new)
Later with this commit in parted 1.9 [2], libparted stopped reporting
those names and reported these instead:
linux-swap(v0)
linux-swap(v1)
Demonstration:
# mkswap /dev/sdc1
Setting up swapspace version 1, size = 1048572 KiB
no label, UUID=a2010834-003d-4bf2-9e94-58383fe20a26
# blkid /dev/sdc1
/dev/sdc1: UUID="a2010834-003d-4bf2-9e94-58383fe20a26" TYPE="swap"
# parted -v
parted (GNU parted) 2.1
...
# parted /dev/sdc unit s print
...
Number Start End Size Type File system Flags
1 2048s 2099199s 2097152s primary linux-swap(v1)
As GParted now requires libparted 2.2 or later [3], remove recognition
for those no longer libparted reported linux-swap names. Note that the
"swap" name is reported by blkid.
[1] http://git.savannah.gnu.org/cgit/parted.git/commit/?id=98a53fd115ca012edf226525b8d4be628454f99e
Enable support for swsusp partitions, and the ability to
differentiate between old and new versions of linux-swap partitions.
Changed the swap_init signature and removed extra ped_geometry_read
from _swap*_open.
[2] http://git.savannah.gnu.org/cgit/parted.git/commit/?id=cfafa4394998a11f871a0f8d172b13314f9062c2
Rationalise linux-swap fs names, and add a "linux-swap" alias
[3] 8df975c7d1
Increase minimum required libparted to 2.2 (!22)
Formatting a partition to cleared over the top of LVM2 PV leaves the
"lvm" flag set on the partition; where as formatting with an actual file
system over the top of an LVM2 PV clears the "lvm" flag. This is true
for both MSDOS and GPT partitioned drives.
Fix by setting the partition type when formatting to cleared too.
Closes!36 - Set partition type when clearing partition contents
Having a member variable named 't' which is used to share state in a
Dialog_Progress object between on_signal_show() and on_cancel() methods
is horrible. Rename to something more meaningful.
Also initialise m_curr_op in the constructor's initialisation list,
rather than later when first used in on_signal_show(). Not strictly
required, but avoids this POD (Plain Old Data) member variable being
undefined in the Dialog_Progress object between construction and when
on_signal_show() previously assigned to it for the first time and
started using it.
* C++ FAQ / Should my constructors use "initialization lists" or
"assignment"?
https://isocpp.org/wiki/faq/ctors#init-lists
Writes the starting device overview information of all known devices to
the top of the saved details HTML. This is so that hopefully we don't
need to additionally ask users for their disk layouts via 'fdisk -l',
'parted print' and 'lsblk' when the saved details file is provided.
Also moves the equals separators "==================" from below to
above each operation so the each section is separated.
Bug 639176 - Save partition layout to gparted_details.htm
Some icon themes only provide large icons for stock items. This can
cause problems like overly large icons appearing in the GParted UI.
Found on Kubuntu 16.04 LTS with default breeze icon theme.
Be compatible with these icon themes by forcing scaling of stock icons
to the requested size.
Icons are used either by Gtk::Image widgets, or Gtk::CellRendererPixbuf
objects for comboboxes/treeviews. For Gtk::Image widgets we add
Utils::mk_image() that constructs Gtk::Image widgets and then sets the
pixel-size property. For Gtk::CellRendererPixbuf we add
Utils::mk_pixbuf() that first loads a Gdk::Pixbuf and then scales if
needed.
Closes#39 - After GTK3 port icons are too big on KDE
To further help in diagnosing root authorisation issues by reporting the
error message to the terminal too. Also set a failure exit status when
terminating with this error.
Example:
$ ./gpartedbin
GParted 0.33.0-git
configuration --enable-online-resize
libparted 3.2
Root privileges are required for running GParted
$ echo $?
1
Closes!34 - Display more version and configuration information
So that the new version and configuration information is displayed even
if the gpartedbin executable is run as a non-root user. To help with
diagnosing root authorisation issues with the gparted shell wrapper
script.
Closes!34 - Display more version and configuration information
For consistency save the 3 same lines of information into the saved
operation details. None of the new 3 lines are subject to translation.
As this information is really for our benefit when supporting users
leaving them as English is OK. Also "GParted" and "Libparted", as
previously used, are proper nouns so they were never changed as part of
any language translation. See with:
egrep -r '"(GParted|Libparted)"' po/
Closes!34 - Display more version and configuration information
So that we might get more information from users when helping them.
Starting GParted from the command line now looks like this:
# ./gpartedbin
GParted 0.33.0-git
configuration --enable-online-resize
libparted 3.2
Closes!34 - Display more version and configuration information
Running GParted on AltLinux with dmraid installed but with no configured
RAID arrays produces this error:
# gparted
...
Could not stat device /dev/mapper/No RAID disks - No such file or directory.
Most distributions use dmraid 1.0.0.rc16 which reports no raid disks
like this:
# dmraid -sa -c
no raid disks
# echo $?
1
However AltLinux had the slightly older version, dmraid 1.0.0.rc14,
which reported no raid disks like this:
# dmraid -sa -c
No RAID disks
# echo $?
0
So because dmraid 1.0.0.rc14 reported success, exit status 0, and the
"No RAID disks" message was not in lower case, that text was considered
a disk device in a DMRaid array. Fix by checking for "no raid disks" in
any case.
Bug 786031 - Could not stat device /dev/mapper/No RAID disks - No such
file or directory
Glibmm has implemented a ustring::compose() set of methods [1] since
Glibmm 2.16, circa 2008. So replace String::ucompose(). Note that
GParted already requires glibmm >= 2.32 as set in configure.ac.
This commit just replaces all the method calls. Edit created by:
sed -i 's|String::ucompose *|Glib::ustring::compose|' src/*.cc
[1] Glibmm Reference Manual, Glib::ustring Class, compose() method
https://developer.gnome.org/glibmm/2.32/classGlib_1_1ustring.html#a64ff7ac3d9e9899c2910f1d831f8d500Closes#46 - Drop compose subdir
Fix to make mkfs.f2fs properly handle labels longer than 16 characters
was included in f2fs-tools 1.2.0 [1]. The oldest supported
distributions now include this release:
Distro EOL f2fs-tools
- Debian 8 2020-Jun 1.4.0
- RHEL / CentOS 7 2024-Jun 1.4.1
- SLES 12 2027-Oct Unknown
- Ubuntu 14.04 LTS 2019-Apr 1.2.0
Note that on Ubuntu 14.04 LTS blkid from util-linux 2.20.1 is too old to
recognise F2FS file systems, as 2.23 is required for F2FS support [2].
mkfs.f2fs claims the maximum label length is less than 512 characters,
but actually accepts 512 characters.
# label=`head -c 1024 < /dev/zero | tr '\0' 'A'`
# mkfs.f2fs -l `echo -n "$label" | cut -c1-513` /dev/sdb10
F2FS-tools: mkfs.f2fs Ver: 1.4.0 (2014-09-18)
Error: Volume Label should be less than 512 characters
Usage: mkfs.f2fs [options] device [sectors]
[options]:
-a heap-based allocation [default:1]
-d debug level [default:0]
-e [extension list] e.g. "mp3,gif,mov"
-l label
-o overprovision ratio [default:5]
-s # of segments per section [default:1]
-z # of sections per zone [default:1]
-t 0: nodiscard, 1: discard [default:1]
sectors: number of sectors. [default: determined by device size]
# echo $?
1
# mkfs.f2fs -l `echo -n "$label" | cut -c1-512` /dev/sdb10
F2FS-tools: mkfs.f2fs Ver: 1.4.0 (2014-09-18)
Info: Label = AAAAAAAAAAAA...[trimmed from 512 "A"s]...AAAAAAAAAAAA
Info: sector size = 512
Info: total sectors = 1048576 (in 512bytes)
Info: zone aligned segment0 blkaddr: 256
Info: Discarding device
Info: This device doesn't support TRIM
Info: format successful
# echo $?
0
# blkid -V
blkid from util-linux 2.25.2 (libblkid 2.25.0, 24-Oct-2014)
# blkid /dev/sdb
/dev/sdb10: LABEL="AAAAAAAAAAAA...[only 127 "A"s]...AAAAAAAAAAAA"
UUID="f47f3fdc-dd91-4616-bb6d-0d643a884265" TYPE="f2fs"
PARTUUID="3bb4bef8-9494-4e82-8dda-5d8edd9c60d9"
As blkid only reports the first 127 characters and is the only command
used for reading the label of an F2FS file system, use this as the new
increased limit.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-tools.git/commit/?id=9799d6364dc93e1fd259d812d4a50ed984a6456b
mkfs: handle labels longer than 16 characters
[2] https://mirrors.edge.kernel.org/pub/linux/utils/util-linux/v2.23/v2.23-ReleaseNotes
"add Flash-Friendly File System (f2fs) support [Alejandro Martinez
Ruiz]"
Closes!29 - Enhance F2FS support
On CentOS 7 with f2fs-tools 1.4.1, checking an F2FS file system fails
like this:
# fsck.f2fs -f -y -a /dev/sdb3
Info: Force to fix corruption
fsck.f2fs: invalid option -- 'y'
Error: Unknown option ?
Usage: fsck.f2fs [options] device
[options]:
-a check/fix potential corruption, reported by f2fs
-d debug level default:0]
-f check/fix entire partition
-t show directory tree [-d -1]
# echo $?
1
Turns out that the '-y' option was not available until f2fs-tools 1.10.0
and is identical to the existing '-f' option anyway [1], which GParted
already uses. Just remove the '-y' option passed to fsck.f2fs.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-tools.git/commit/?id=55ee9e7202f84168f868d863da8ed1c4995a0e6d
fsck.f2fs: add -y for generic fsck
Closes!29 - Enhance F2FS support
Before this commit [1] first included in f2fs-tools 1.5.0, dump.f2fs
didn't report the total space used by the file system. This causes F2FS
file system usage not be reported on older distributions, including
RHEL/CentOS 7 and Debian 8.
On CentOS 7:
# rpm -q f2fs-tools
f2fs-tools-1.4.1-2.el7.nux.x86_64
# dump.f2fs -d 1 /dev/sdb3 | egrep 'sector size =|total.*sectors ='
Info: sector size = 512
Info: total sectors = 2097152 (in 512 bytes)
On Fedora 28:
# rpm -q f2fs-tools
f2f2-tools-1.10.0-1.fc28.x86_64
# dump.f2fs -d 1 /dev/sdb2 | egrep -a 'sector size =|total.*sectors ='
Info: sector size = 512
Info: total sectors = 3145728 (1536 MB)
Info: total FS sectors = 2621440 (1280 MB)
"total sectors" reports the size of the partition.
"total FS sectors" reports the size of the file system.
Cope with the file system size being missing. Pass -1 as the file
system size to partition.set_sector_usage() in this case. Note that
this means GParted won't be able to detect and report unallocated space
within the partition when using f2fs-tools < 1.5.0.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-tools.git/commit/?id=fea6162a6db9af527648262d9fbb0335a0cc5460
fsck.f2fs: show total sectors consumed by filesystem
Closes!29 - Enhance F2FS support
On Fedora 28 with f2fs-tools 1.10.0, dump.f2fs is producing NUL
characters in it's output and this completely breaks the parsing code in
f2fs::set_used_sectors(). Glib::Regex, as used by
Utils::regexp_label(), just doesn't match any text after the first NUL
character from the output.
# dump.f2fs -d 1 /dev/sdb1
Info: Debug level = 1
Info: [/dev/sdb1] Disk Model: VBOX HARDDISK 1.0 ^@^@^@^@^@^@^@^...
Info: Segments per section = 1
Info: Sections per zone = 1
Info: sector size = 512
Info: total sectors = 2097152 (1024 MB)
...
Grep thinks the output is binary too:
# dump.f2fs -d 1 /dev/sdb1 | \
> egrep 'valid_block_count|user_block_count|log_blocksize|sector size =|total FS sectors ='
Binary file (standard input) matches
# dump.f2fs -d 1 /dev/sdb1 | \
> egrep --text 'valid_block_count|user_block_count|log_blocksize|sector size =|total FS sectors ='
Info: sector size = 512
log_blocksize [0x c : 12]
Info: total FS sectors = 2097152 (1024 MB)
user_block_count [0x 36400 : 222208]
valid_block_count [0x 2 : 2]
Re-write set_used_sectors() using string find() and sscanf() to be
similar to how a number of the other set_used_sectors() are written for
other file systems.
Closes!29 - Enhance F2FS support
Now that GParted requires Gtk3 there is no need to use floating point
numbers for compatibility with Gtk <= 2.22. Replace with symbolic
alignment constants.
Relevant commit history:
* 6efa623401
Add optional yalign argument to Utils::mk_label() method
* be2689ad25
Stop using deprecated widget alignment enumerators (#652044)
btrfs-progs 3.12 includes 'btrfs filesystem label /dev/PTN NEWLABEL'
functionality so stop checking for this before enabling setting the
label.
$ btrfs version
Btrfs v3.12
$ btrfs filesystem label --help
usage: btrfs filesystem label [<device>|<mount_point>] [<newlabel>]
Get or change the label of a filesystem
With one argument, get the label of filesystem on <device>.
If <newlabel> is passed, set the filesystem label on <newlabel>.
$ echo $?
0
Worst case scenario is that some how an old version of the btrfs command
is used which doesn't support the labelling functionality. Then this
commit would change GParted from disallowing labelling of a btrfs, to
allowing it, but presumably it would fail with an error from the btrfs
command reporting so. Arguably better from a support point of view.
Closes!26 - Remove support for btrfs-progs < 3.12
This commit [1] from btrfs-progs 3.12 removed the previously deprecated
programs btrfsctl and btrfs-show. As btrfs-progs 3.12 is now the
minimum requirement, remove support for those removed programs.
This commit is just removing the use of btrfs-show as a fallback to read
the label.
Note that 'btrfs-show /dev/PTN' didn't distinguish between a label of
"none" and no label. Hence the logic in btrfs::read_label() to do with
matching the label "none", or matching the label with or without single
quotes. Unfortunately as identified in this commit [2]
'btrfs filesystem show /dev/PTN' is subject to the same issue, but only
when the file system is mounted and only for btrfs-progs 3.12. This was
fixed by this commit [3] from btrfs-progs 3.14.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/kdave/btrfs-progs.git/commit/?id=f243fcd1b2aa55ffadfbcc032c66dedbee56e79e
Removing btrfsctl, btrfs-vol, btrfs-show
[2] eca732fb0c
Update parsing of btrfs filesystem show for the label (#733601)
[3] https://git.kernel.org/pub/scm/linux/kernel/git/kdave/btrfs-progs.git/commit/?id=a156b967ed9bd606afa8dc402451abcf07226c17
btrfs-progs: make filesystem show by label work
Closes!26 - Remove support for btrfs-progs < 3.12
PATCHSET OVERVIEW
The oldest supported distributions have these versions of the Linux
kernel and btrfs-progs:
Distro EOL kernel btrfs-progs
- RHEL / CentOS 7 2024-Jun 3.10.0 4.9.1
- Ubuntu 14.04 LTS 2019-Apr 4.4.0 3.12
- Debian 8 2020-Jun 3.16.0 3.17
- SLES 12 2027-Oct 3.12.28 3.16
Making the oldest supported packages be kernel 3.10 and btrfs-progs 3.12
allows the btrfs support code to be simplified by removing backward
compatibility.
THIS CHANGE
Remove old workaround for ignoring the error when resizing a btrfs to
the same size on Linux kernel < 3.2.
Also now that only exit status 0 is considered successful from btrfs
resize, the EXEC_CHECK_STATUS flag to execute_command() can be used,
rather than having to separately call set_status() afterwards.
Relevant commit history:
* 11d044dba0
Don't ignore any errors resizing btrfs on Linux >= 3.2 (#669389)
* a580abbc30
Use newer btrfs multi-tool control command first
Closes!26 - Remove support for btrfs-progs < 3.12
The title has never been set in this case, and defaulted to the name of
the executable 'gpartedbin'. Fix this.
Closes#44 - Title not set in Resize/Move dialog for extended partitions
'launched' local POD (Plain Old Data) variable was left uninitialised,
but was set in both the try and catch clauses. Best practice is to
initialise when defined, so do that instead. Cosmetic change.
It is not creating a dialog (a pop-up window managed by GParted code
itself). It is launching independent yelp program to display the help,
so remove the "_dialog" from the name to avoid any possible confusion.
Originally, if the yelp command was not installed, attempting to display
help produced an error dialog with this message:
Failed to execute child process "yelp" (No such file or directory)
However since this commit during the Gtk 3 port [1] the error message
became this less useful one:
Operation not supported
Two attempts are made to display the GParted Manual, first using
gtk_show_uri() and second by executing the yelp command directly. Prior
to the aforementioned commit [1] both methods returned the failure
reason using the same 'error' variable. Hence reported the message
"Failed to execute child process "yelp" ..." from the second attempt.
However that commit had to re-code the second method as part of the Gtk
3 port and use a different error returning mechanism, thus the use of
different variable 'e'. But the dialog was left reporting the message
from the original 'error' variable, thus reporting "Operation not
supported" message from the first attempt using gtk_show_uri().
Fix by again displaying the message from the second failure into the
error dialog. Also make it very clear there are two error returning
variables by naming them 'error1' and 'error2_msg'.
[1] 2953778a4c
port-to-gtk3: Use Gdk::AppLaunchContext to launch yelp (#7)
Update GParted to specify the GParted Manual using the new GNOME 3 way
with the 'help:' prefix to avoid yelp reporting this error:
Document Not Found
The URI 'ghelp:gparted' does not point to a valid page.
Closes!24 - Port to GNOME 3 yelp-tools documentation infrastructure
According to the GIT history the lines were added by this commit:
8d808c0b62
gparted-0.3.6 - code recreation from Source Forge
Looking at the SVN history this commit actually fleshed out the
implementations of fat16::get_label() and fat32::get_label() and added
the commented #includes:
https://sourceforge.net/p/gparted/svn/118
Added read label support for fat16 and fat32 using mtools mlabel command
2008-02-12
Then this SVN commit moved the mtools temporary file handling code into
Utils.cc, leaving behind the commented #includes:
https://sourceforge.net/p/gparted/svn/124
Added MTools temporary file handling functions
2008-02-19
Finally this commit removed fat32.cc by merging the code with fat16.cc:
519af1a7c0
Combine duplicate code for fat16/32
So remove the left behind commented #includes from fat16.cc.
A forum user had a case where they wanted to grow their in use root,
ext4 file system. GParted supports this, but the partition was a
logical partition inside an extended partition and GParted doesn't
support resizing an extended partition while any contained logical
partitions are busy.
Example layout:
Partition File System Mount Point
/dev/sdb1 ntfs
/dev/sdb2 [busy]
/dev/sdb5 [busy] ext4 /
unallocated unallocated
So just allow extended partitions to be resized online when online
partition resizing is available via libparted.
NOTE:
The block device that the Linux kernel provides for an extended
partition just maps to the first 1 KiB of the extended partition where
the Extended Boot Record is stored, and does not include any of the
contained logical partitions. Therefore no application can care that
the extended partition is resized while a logical partition is in use
because it can't use the extended partition block device to access any
data.
The on disk layout looks like this:
# fdisk -l /dev/sdb
Disk /dev/sdb: 8589 MB, 8589934592 bytes, 16777216 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x0007650e
Device Boot Start End Blocks Id System
/dev/sdb1 2048 1050623 524288 7 HPFS/NTFS/exFAT
/dev/sdb2 1050624 2101247 525312 5 Extended
/dev/sdb5 1052672 2101247 524288 83 Linux
# parted /dev/sdb unit s print free
Model: ATA VBOX HARDDISK (scsi)
Disk /dev/sdb: 16777216s
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Disk Flags:
Number Start End Size Type File system Flags
63s 2047s 1985s Free Space
1 2048s 1050623s 1048576s primary ntfs
2 1050624s 2101247s 1050624s extended
5 1052672s 2101247s 1048576s logical ext4
2101248s 16777215s 14675968s Free Space
The kernel's partition sizes from /sys/block/sdb/sdb${N}/{start,size}
shows extended partition 2 has a size of only 2 sectors:
# for N in 1 2 5
> do
> echo -e "/dev/sdb${N}\tstart=`cat /sys/block/sdb/sdb${N}/start`\tsize=`cat /sys/block/sdb/sdb${N}/size`"
> done
/dev/sdb1 start=2048 size=1048576
/dev/sdb2 start=1050624 size=2
/dev/sdb5 start=1052672 size=1048576
The EBR read from the whole of extended partition 2 block device:
# hexdump -C /dev/sdb2
00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
000001b0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 86 |................|
000001c0 06 41 83 cb 09 82 00 08 00 00 00 00 10 00 00 00 |.A..............|
000001d0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
000001f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.|
00000200 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
00000400
Closes!23 - Enable online resizing of extended partitions
With the Gtk3 port the File System Support dialog has become too wide
because the legend text is no longer wrapped. Set the max-width-chars
property to specify the natural size of the widget in terms of
characters [1]. It is converted to pixels using the average character
width in the current font.
Also use PACK_EXPAND_WIDGET when adding the label to the box so that if
the dialog is resized extra space is used to increase the size of this
child widget [2].
[1] GNOME HowDoI / Labels
https://wiki.gnome.org/HowDoI/Labels
[2] Gtkmm 3.0 Enums and Flags, enum Gtk::PackOptions
"PACK_EXPAND_WIDGET Space is expanded, with extra space filled by
increasing the child widget size."
https://developer.gnome.org/gtkmm/3.0/group__gtkmmEnums.html#ga83727a1b6fed51566dfd5c8e58890dbaCloses#7 - Port to Gtk3
In Gtk2 the up and down buttons in a SpinButton were smaller leaving
space for 7 digits before scrolling the entry. In Gtk3 the up and down
buttons are much larger leaving only space for 4 digits. This occurs in
the SpinButtons in the Dialog_Base_Partition class as displayed in the
New Partition, Paste and Resize/Move dialogs.
Set width-chars property of all Gtk::SpinButtons to ensure 7 digits can
be displayed before scrolling the entry.
Closes#7 - Port to Gtk3
In Gtk3 the progress bar height is fixed and defined by the CSS theme in
use. Changing the widget allocation size does nothing, it is always
rendered the same way.
In many themes, including Adwaita, the progressbar is very, very thin.
Provide custom CSS to specify a height of 8 pixels.
The CSS source string has to be differentiated for Gtk pre and post
3.20, because Gtk 3.20 introduced some breaking changes in the way CSS
is handled.
References:
[1] Migrating from GTK+ 2.x to GTK+ 3 - Parsing of custom resources
https://developer.gnome.org/gtk3/stable/gtk-migrating-GtkStyleContext-parsing.html
[2] Gtk3 Reference Documentation - Changes in GTK+ 3.20
https://developer.gnome.org/gtk3/stable/ch32s10.html
[3] Gnome/HowDoI - Custom Style
https://wiki.gnome.org/HowDoI/CustomStyleCloses#7 - Port to Gtk3
The pulsebar looks very small and needs to be widened. The pulsebar is
packed inside the statusbar so that it displays activity text on the
left side and the pulsebar on the right side. Ideally we want the space
to be evenly divided for the textual messages and for the pulsebar
activity indicator.
For this we just have to set the 'homogeneous' property to TRUE for the
statusbar (note that GtkStatusBar inherits from GtkBox).
Also vertically align the pulsebar to the center of the statusbar. This
is achieved setting the 'valign' property to Gtk::ALIGN_CENTER for the
pulsebar widget.
Closes#7 - Port to Gtk3
There is a bug affecting Gtk+ 3.22.8 to 3.22.30 in which destroying a
GtkComboBox when it is not hidden results in this message:
Gtk-CRITICAL **: gtk_widget_is_drawable: assertion 'GTK_IS_WIDGET (widget)' failed
This happens in GParted when some dialogs are closed, for example the
Create New Partition and Create Partition Table dialogs. To work around
the issue we call Gtk::Dialog::hide() in the destructors of our dialog
classes.
The issue was fixed in Gtk 3.24.0.
* Gtk 3.22.8 was released in February 2017.
* Gtk 3.24.0 was released in September 2018.
References:
[1] Gtk Issue - GtkComboBox::private::popup_window can be NULL
https://gitlab.gnome.org/GNOME/gtk/issues/125
[2] Gtk commit - combobox: popdown() the menu during unmap()
7401794de6
[3] Gtk commit - Check for NULL priv->popup_window in
gtk_combo_box_popdown()
aa5d926c84Closes#7 - Port to Gtk3
There is a bug in Gtkmm3 when setting accelerator keys on a
Gtk::MenuItem, the accelerator keys work but are not displayed when the
menu is drawn. This happens for Gtk::MenuItems, including derived
objects, that are constructed with a non-default constructor.
All non-default constructors of Gtk::MenuItem, and subclasses, work by
creating themselves a Gtk::AccelLabel and packing it inside the menu
item. But in Gtk3 GtkMenuItem are created with a GtkAccelLabel already
packed in as a child and that accel label should be used instead.
To workaround the issue we only use the default constructor for
Gtk::MenuItem and subclasses. This is easy to do because we only have
to change the wrappers in MenuHelpers.cc.
This bug affects Gtkmm version 3.0.0 to 3.22.2 and was fixed in
Gtkmm 3.22.3.
* Gtkmm 3.0.0 was released in April 2011
* Gtkmm 3.22.3 was released in November 2018
References:
[1] Bug Report on the Gtkmm mailing list
https://mail.gnome.org/archives/gtkmm-list/2018-February/msg00006.html
[2] Commit - Gtk::MenuItem: Fix add_accel_label()
e5c8c2df67Closes#7 - Port to Gtk3
Now GParted compiles with Gtkmm3, but we get a failed assertion doing
the following:
* Select a device with more than 1 partition
* Select a partition
* Activate 'Refresh Devices' (or do any operation that causes it, like
mount/unmount etc.)
This is the failed assertion:
**
ERROR:Win_GParted.cc:1152:void GParted::Win_GParted::set_valid_operations(): assertion failed: (valid_display_partition_ptr( selected_partition_ptr ))
Aborted (core dumped)
Where is the problem?
Win_GParted::Refresh_Visual() calls TreeView_Detail::load_partitions()
to clear and refill the treeview.
The problem is in GParted::TreeView_Detail::load_partitions() at
TreeView_Detail.cc:91:
treestore_detail->clear();
This activates TreeView_Detail::on_selection_changed() which in turn
activates Win_GParted::on_partition_selected() passing an old, stale
pointer as an argument. This triggers the failed assertion.
Why does this happen with Gtk3 and not with Gtk2?
First a bit of background of GtkTreeView:
What happens to the selection in a GtkTreeView when the selected row
is removed?
With Gtk2 the selection simply becomes empty, so nothing is selected
afterwards. With Gtk3 this was changed [1] and selection moves to an
adjacent row.
gtk_tree_store_clear() removes rows one by one. While removing rows the
selection changed signal is emitted. With Gtk2 it is emitted only one
time, to indicate that selection has become empty. With Gtk3 it is
instead emitted several times, each time indicating that selection has
moved to the adjacent row.
The handler TreeView_Detail::on_selection_changed() only takes action
when the selection is not empty. So with Gtk3 it really takes action
and activates Win_GParted::on_partition_selected() with a pointer to old
data.
What's the purpose of TreeView_Detail::on_selection_changed()?
Its task is to update the selection in the drawing area above the
TreeViewDetail, the DrawingAreaVisualDisk, so that the selected
partition stays in sync on the two widgets.
Fix by blocking the signal handler during the treeview clear.
Reference:
[1] Commit - treeview: Handle the case where the cursor row gets deleted
1a2932ba29Closes#7 - Port to Gtk3
gdk_spawn_command_line_on_screen() is not present in Gtk3. The
documentation from Gtkmm 2.24 states [1]:
gdk_spawn_command_line_on_screen has been deprecated since version
2.24 and should not be used in newly-written code. This function is
being removed in 3.0. Use either g_spawn_command_line_sync(),
g_spawn_command_line_async() or GdkAppLaunchContext instead.
g_spawn_command_line_sync() and g_spawn_command_line_async() are screen
/ display agnostic, as such we would loose functionality. There is a
workaround, which involves setting the DISPLAY environment variable [2],
but it's a weak solution (and I don't know if it works on backends
other than X11).
GdkAppLaunchContext is an implementation of GIO's GAppLaunchContext that
handles launching an application in a graphical context [3]. Therefore
use GdkAppLaunchContext and GIO's GAppInfo.
GdkAppLaunchContext was introduced in Gtk2 version 2.14. The C++
wrapper Gdk::AppLaunchContext was introduced only in Gtkmm3 version 3.4
[4]. Bump the minimum required version of Gtkmm to 3.4.0 for this
requirement.
GAppInfo was introduced in GLib version 2.16. The C++ wrapper
Gio::AppInfo was introduced in Giomm version 2.16. Note that the
minimum required version for glibmm is already 2.32.
[1] GDK 2 Reference Manual, GdkScreen, gdk_spawn_on_screen()
https://developer.gnome.org/gdk2/2.24/GdkScreen.html#gdk-spawn-on-screen
[2] Migrating from GTK+ 2.x to GTK+ 3 - "Use GIO for launching applications"
https://developer.gnome.org/gtk3/stable/gtk-migrating-2-to-3.html#id-1.6.3.3.7
[3] GDK 3 Reference Manual - "Application launching"
https://developer.gnome.org/gdk3/stable/gdk3-Application-launching.html
[4] Gtkmm 3.4 Gdk::AppLaunchContext Class Reference, Detailed Description
https://developer.gnome.org/gtkmm/3.4/classGdk_1_1AppLaunchContext.html#detailsCloses#7 - Port to Gtk3
GParted uses Gtk::TreeViewColumn::get_cell_renderers(). This is not
present in Gtkmm3. Now Gtk::TreeViewColumn inherits from
Gtk::CellLayout and we have to use Gtk::CellLayout::get_cells() instead.
GtkCellLayout was introduced in Gtk2 version 2.18 as the common
interface for containers of cell renderers.
The C++ wrapper Gtk::CellLayout was introduced in Gtkmm2 version 2.18,
but Gtk::TreeViewColumn was never made to inherit from Gtk::CellLayout
to avoid breaking the API / ABI. That change was made for Gtkmm3.
This is an excerpt from gtkmm/treeviewcolumn.h header in Gtkmm2:
// TODO: Should be deprecated, but we cannot derive from CellLayout
// without breaking API and ABI.
/** Returns a list of all the cell renderers in the column,
* in no particular order.
*
* @return A list of Gtk::CellRenderers.
*/
Glib::ListHandle<CellRenderer*> get_cell_renderers();
Replace Gtk::TreeViewColumn::get_cell_renderers() with base class
method Gtk::CellLayout::get_cells().
Reference:
[1] Commit - "Deprecate get_cell_renderers implementations"
6abc52a29dCloses#7 - Port to Gtk3
In Gtk2 widgets draw themselves in response to the expose event signal.
In Gtk3 widgets draw themselves in response to the GtkWidget::draw
signal, and the signal handler gets a Cairo context as an argument.
Convert Gtk::DrawingArea rendering code to respond to the
GtkWidget::draw signal.
This commit is specific to the drawing area in the Create new Partition
dialog and the Resize/Move dialog.
Reference:
[1] Migrating from GTK+ 2.x to GTK+ 3 - "The GtkWidget::draw signal":
https://developer.gnome.org/gtk3/stable/ch26s02.html#id-1.6.3.4.11Closes#7 - Port to Gtk3
In Gtk2 widgets draw themselves in response to the expose event signal.
In Gtk3 widgets draw themselves in response to the GtkWidget::draw
signal, and the signal handler gets a Cairo context as an argument.
Convert Gtk::DrawingArea rendering code to respond to the
GtkWidget::draw signal.
This commit is specific to the drawing area contained in the main
application window (also called the DrawingAreaVisualDisk).
Reference:
[1] Migrating from GTK+ 2.x to GTK+ 3 - "The GtkWidget::draw signal":
https://developer.gnome.org/gtk3/stable/ch26s02.html#id-1.6.3.4.11Closes#7 - Port to Gtk3
In Gtk2 widgets draw themselves in response to the expose event signal.
In Gtk3 widgets draw themselves in response to the GtkWidget::draw
signal, and the signal handler gets a Cairo context as an argument.
Convert Gtk::DrawingArea rendering code to respond to the
GtkWidget::draw signal.
This commit is specific to the drawing area in the Partition Info
dialog.
Reference:
[1] Migrating from GTK+ 2.x to GTK+ 3 - "The GtkWidget::draw signal":
https://developer.gnome.org/gtk3/stable/ch26s02.html#id-1.6.3.4.11Closes#7 - Port to Gtk3
Starting from Gtkmm3 Gdk::Cursor objects cannot be constructed directly,
but instead you have to get a smart pointer to an instance by calling
the static member function Gdk::Cursor::create().
Gdk::Cursor::create() returns a Glib::RefPtr<Gdk::Cursor> object.
Gtkmm3 always uses Glib::RefPtr<Gdk::Cursor> in its interface and never
plain Gdk::Cursor.
Reference:
[1] Programming with gtkmm3, Changes in gtkmm3:
https://developer.gnome.org/gtkmm-tutorial/3.24/changes-gtkmm3.html.en
"... Gdk::Cursor are now used via Glib::RefPtr."
Closes#7 - Port to Gtk3
In Gtk3 individual headers cannot be included directly in application
code, only the header <gtk/gtk.h> can be included (with a few exceptions
for some platform specific headers).
This has always been considered good practice even for Gtk2, though was
not a hard requirement.
In Gtk3 this is enforced by preprocessor checks. Failure to do so
yields a preprocessor error and compilation fails:
"error: Only <gtk/gtk.h> can be included directly."
Change specific Gtk header includes to:
#include <gtk/gtk.h>
References:
[1] Migrating from GTK+ 2.x to GTK+ 3 - "Do not include individual headers"
https://developer.gnome.org/gtk3/stable/gtk-migrating-2-to-3.html#id-1.6.3.3.3
[2] Commit - "Remove all traces of GDK_PIXBUF/GTK_DISABLE_SINGLE_INCLUDES"
5e29973773Closes#7 - Port to Gtk3
Now that we are compiling against Gtkmm3 there are missing declarations
of Gtkmm identifiers due to changes in Gtkmm internal header structure.
All we have to do is bring back the declarations by including the
appropriate headers where needed.
Add necessary Gtkmm header includes.
Closes#7 - Port to Gtk3
Now that we are compiling against Gtkmm3 there are missing declarations
of Glibmm identifiers due to changes in Gtkmm internal header structure.
All we have to do is bring back the declarations by including the
appropriate headers where needed.
Add necessary Glibmm header includes.
Closes#7 - Port to Gtk3
The Gtk::Widget::modify_fs() API was removed in Gtkmm3 [1] and also
there is no direct replacement. GParted uses this in one place. So
instead use the C gtk_widget_modify_fg() version that is still present
in Gtk3.
This is just a temporary change to port GParted to Gtk3. In future this
will be replaced as part of the switch from Gdk::Color to Gtk::RGBA,
since Gdk::Color was deprecated in Gtkmm 3.10 [2].
Reference:
[1] ee432e2190
commit message "... Remove the modify_*() methods, ..."
[2] Gtkmm 3.10 Gdk::Color Class Reference
https://developer.gnome.org/gtkmm/3.10/classGdk_1_1Color.html#details
Deprecated:
Use Gdk::RGBA instead.
Closes#7 - Port to Gtk3
GParted uses Gtk::Menu_Helpers::MenuList helper class to access
individual menu items. This helper class made it easy to navigate menu
items by index.
Gtk::Menu_Helpers::MenuList was removed in the switch from Gtkmm2 to
Gtkmm3 [1]. Instead, use a separate std::map<Gtk::MenuItem*> to keep
track of individual Gtk::MenuItem objects.
Reference:
[1] Gtkmm 3 commit "MenuShell: Remove items()." removed the code
c8e47b0db5Closes#7 - Port to Gtk3
Gtk::Menu_Helpers::Element class and subclasses help in Gtk::MenuItem
widgets and also automate tasks like registering keyboard accelerators
when parented to a top-level window [1][2].
Gtk::Menu_Helpers::Element class and subclasses were removed in Gtkmm3
[3]. Provide compatible implementations under the namespace
GParted::Menu_Helpers.
References:
[1] gtkmm: Gtk::Menu_Helpers Namespace Reference
https://developer.gnome.org/gtkmm/2.24/namespaceGtk_1_1Menu__Helpers.html
[2] gtkmm: Gtk::Menu_Helpers::Element Class Reference
https://developer.gnome.org/gtkmm/2.24/classGtk_1_1Menu__Helpers_1_1Element.html
[3] Gtkmm 3 commit "MenuShell: Remove items()." removed the code
c8e47b0db5Closes#7 - Port to Gtk3
During the switch from Gtk2 to Gtk3 keyname constants were renamed from
GDK_<KeyName> to GDK_KEY_<KeyName> [1]. This was done to avoid name
clashes in gobject-introspection and language bindings. The new
constant names were also backported to Gtk 2.22 [2].
Make use of the new constant names.
References:
[1] Migrating from GTK+ 2.x to GTK+ 3 - "Replace GDK_<keyname> with GDK_KEY_<keyname>"
https://developer.gnome.org/gtk3/stable/gtk-migrating-2-to-3.html#id-1.6.3.3.6
[2] Commit - "gdk: Prefix keys with _KEY by default"
750c81f43dCloses#7 - Port to Gtk3
Originally in Gtk2, by default, dialogs showed a horizontal separator
between the content area and the action area (buttons). GParted
explicitly called Gtk::Dialog::set_has_separator(false) for all it's
dialogs to remove the separator.
In Gtk2/Gtkmm2 2.22, separators were deprecated [1]. In Gtk3/Gtkmm3
separators in dialogs were removed altogether, including all the related
APIs [2][3][4]. Therefore remove all calls.
References:
[1] Commit - "Document separator-related api in GtkDialog as deprecated"
6f6650e6cf
[2] Commit - "Remove separators from dialogs"
d433a60611
[3] GTK+ 3.0.0 NEWS file
https://gitlab.gnome.org/GNOME/gtk/blob/3.0.0/NEWS#L779
"
Overview of Changed from GTK+ 2.90.6 to 2.90.7
==============================================
* Various deprecated APIs have been removed:
...
- GtkDialog separators, including the GtkDialog::has-separator
property, including setter/getter, the GTK_DIALOG_NO_SEPARATOR
flag and the GtkMessageDialog::use-separator style property
"
[4] Gtkmm 3.0.0 NEWS file
https://gitlab.gnome.org/GNOME/gtkmm/blob/3.0.0/NEWS#406
"
* Dialog:
- Remove get/set_has_separator() and property.
- Constructors: Remove use_separator parameters.
"
Closes#7 - Port to Gtk3
Use of Glib::thread_init() was deprecated in glibmm 2.32 [1]. The
oldest supported distributions have these versions:
Debian 8 glibmm 2.42.0
RHEL / CentOS 7 glibmm 2.56.0
SLES 12 glibmm 2.38.1
Ubuntu 14.04 LTS glibmm 2.39.93
Checking further the glibmm 2.32 reference manual says this about
Glib::thread_init() [2]:
Initializes the GLib thread system.
Deprecated:
Calling thread_init() is no longer necessary and no longer has any
effect.
However only some of the glibmm example programs had Glib::thread_init()
removed, others had it replaced by Glib::init() [3]. Again the glibmm
2.32 reference manual says this about Glib::init() [4]:
Initialize glibmm.
You may call this more than once. You do not need to call this if
you are using Glib::MainLoop or Gtk::Main, because they call it for
you.
GParted does call Gtk::Main and test_PipeCapture does call
Glib::MainLoop. Therefore just raise the minimum version to glibmm 2.32
and remove both calls to Glib::thread_init().
[1] Glibmm 2.32 NEWS file
https://gitlab.gnome.org/GNOME/glibmm/blob/2.32.0/NEWS#L207
[2] glibmm 2.32, glibmm: Glib Namespace Reference, Glib::thread_init()
https://developer.gnome.org/glibmm/2.32/namespaceGlib.html#ab26d01c776801f1fff00753e97af4fc7
[3] glibmm commit "Avoid use of deprecates API in tests and examples."
3e0fbb22c0
[4] glibmm 2.32, glibmm: Glib Namespace Reference, Glib::init()
https://developer.gnome.org/glibmm/2.32/namespaceGlib.html#ac90aee10d0b90e3d8a96a86b5394f87bCloses!22 - Increase minimums to libparted 2.2 and glibmm 2.32
Raise the minimum required version of GNU Parted from 1.7.1 to 2.2,
released 2010-02-16 [1][2]. The oldest supported distributions, also
with gtkmm >= 2.24, since commit [3], are:
Debian 8 parted 3.2
RHEL / CentOS 7 parted 3.1
SLES 12 parted 3.1
Ubuntu 14.04 LTS parted 2.3
Raising the minimum required version allows removal of optional code
associated with these definitions:
* USE_LIBPARTED_LARGE_SECTOR_SUPPORT
Fallback code reporting ignored device with logical sector size
other than 512 bytes.
* ENABLE_PT_REREAD_WORKAROUND
Fallback code re-attempting to inform the kernel of partition
changes.
[1] GNU Parted 2.2 release announcement
http://lists.gnu.org/archive/html/info-gnu/2010-02/msg00016.html
[2] NEWS file from GNU Parted 2.2
http://git.savannah.gnu.org/cgit/parted.git/tree/NEWS?h=v2.2
[3] 8b42bab1ee
modern-gtk2: Require Gtkmm version 2.24 (!17)
Closes!22 - Increase minimums to libparted 2.2 and glibmm 2.32
The code inconsistently uses GParted:: scope in front of STAT_*.
$ fgrep 'GParted::STAT_' src/*.cc | wc -l
3
$ egrep '[^:]STAT_' src/*.cc | wc -l
41
GParted:: scope resolution is unnecessary as all the code is inside the
GParted scope, except for main(). So remove it.
Closes!20 - Minor namespace and scope operator tidy-ups
The code inconsistently uses GParted:: scope in front of TYPE_*.
$ fgrep 'GParted::TYPE_' src/*.cc | wc -l
35
$ egrep '[^:]TYPE_' src/*.cc | wc -l
83
GParted:: scope resolution is unnecessary as all the code is inside the
GParted scope, except for main(). So remove it.
Closes!20 - Minor namespace and scope operator tidy-ups
The code inconsistently uses GParted:: scope in front of FS_*.
$ fgrep 'GParted::FS_' src/*.cc | wc -l
41
$ egrep '[^:]FS_' src/*.cc | wc -l
441
GParted:: scope resolution is unnecessary as all the code is inside the
GParted namespace, except for main(). So remove it.
Closes!20 - Minor namespace and scope operator tidy-ups
The code inconsistency uses GParted::FS::* and FS::*.
$ fgrep 'GParted::FS::' src/*.cc | wc -l
97
$ egrep '[^:]FS::' src/*.cc | wc -l
152
GParted:: scope resolution is unnecessary as all the code is inside the
GParted namespace, except for main(). So remove it.
Closes!20 - Minor namespace and scope operator tidy-ups
All the other modules are in the GParted namespace, except for main()
which has to be in the global namespace, so put these in the GParted
namespace too.
Closes!20 - Minor namespace and scope operator tidy-ups
Create an active Linux Software RAID member which is larger than /dev
virtual file system and GParted will report the usage figure of the /dev
virtual file system for the SWRAID member.
# df -h /dev
Filesystem Size Used Avail Use% Mounted on
devtmpfs 732M 0 732M 0% /dev
# sgdisk -n 1:1M:+1G /dev/sdb
# mdadm --create --verbose /dev/md1 --level=linear --raid-devices=1 --force /dev/sdb1
mdadm: Defaulting to version 1.2 metadata
mdadm: array /dev/md1 started.
GParted reports the usage of /dev/sdb1 as:
Partition Mount Point Size Used Unused Unallocated
/dev/sdb1 /dev/md1 1.00GiB 0.00B 731.04MiB 292.96MiB
However GParted should have reported the usage as "---" for unknown
because it isn't coded to query the size of the SWRAID member within a
partition.
The fault has been bisected to this commit:
Extend un/mounting and usage reporting to unsupported file systems (!13)
95903efb1f
What happens for busy Linux Software RAID array members:
* GParted_Core::is_busy()
has custom code to identify busy members.
* GParted_Core::set_mountpoints()
has custom code to add the array device name as the "mount point" of
the member.
* GParted_Core::set_used_sectors()
falls into the else not a supported file system (because SWRAID
doesn't have a derived FileSystem implementation class).
* GParted_Core::mounted_set_used_sectors()
is called to get the file system usage of mounted, but unsupported
file systems, such as UFS and any others.
* Utils::get_mounted_filesystem_usage()
is called which queries the kernel using statvfs() and gets the file
system usage of the /dev virtual file system because the array
device name will always start /dev.
Fix by ensuring that GParted only asks the kernel for the usage of paths
which it knows are mount points of mounted file systems. (As read from
/proc/mounts and cached in the Mount_Info module). Also rename the
method, by inserting "_fs", to mounted_fs_set_used_sectors() to remind
us that it is for mounted *file systems* only.
Closes#27 - GParted may report incorrect usage for SWRAID partitions
instead of unknown
S
Third commit in a series to convert Gdk::GC based drawing to Cairo based
drawing. This specific commit makes the transition for the graphical
partition visualizer widget that is used in the main application window.
Closed!17 - Gtk2 modernisation
Second commit in a series to convert Gdk::GC based drawing to Cairo
based drawing. This specific commit makes the transition for the
graphical partition info widget that is used in the "Information about"
dialog.
Closes!17 - Gtk2 modernisation
C++ initialises static member variables before main() is called.
Therefore the static members of:
struct Slots
{
static Gtk::TreeModelColumn<Glib::ustring> text;
static Gtk::TreeModelColumn<bool> sensitive;
private:
static Gtk::TreeModel::ColumnRecord record_;
};
are constructed before Gtk::Main() is called in main(). However the
Gtkmm documentation specifically says that they must be constructed
afterwards [1].
Resolve this by using the Construct On First Use Idiom [2] to delay
initialisation until the slots are first used. Normally this idiom uses
static local objects, however it is being applied to class static
objects here because the objects are accessed in many methods. The
downside of this approach is that the objects are never destructed,
which memory analysers like Valgrind could see as a memory leak, but
that is actually deliberate. That leak can be removed once we can use
C++11 and std::unique_ptr.
[1] gtkmm: Gtk::TreeModelColumnRecord Class Reference
https://developer.gnome.org/gtkmm/2.24/classGtk_1_1TreeModelColumnRecord.html#details
"Neither TreeModel::ColumnRecord nor the TreeModelColumns contain
any real data - they merely describe what C++ type is stored in
which column of a TreeModel, and save you from having to repeat that
type information in several places.
Thus TreeModel::ColumnRecord can be made a singleton (as long as you
make sure it's instantiated after Gtk::Main), even when creating
multiple models from it.
"
[2] C++ FAQ / How do I prevent the "static initialization order
problem"?
https://isocpp.org/wiki/faq/ctors#static-init-order-on-first-useCloses!17 - Gtk2 modernisation
Final part in a series of commits to replace Gtk::OptionMenu widgets
with GParted::OptionComboBox.
This specific commit renames the signal handler callback to match the
previously renamed combobox widget variable names.
Closes!17 - Gtk2 modernisation
Third part in a series of commits to replace Gtk::OptionMenu widgets
with GParted::OptionComboBox.
This specific commit is about file system combobox.
Closes!17 - Gtk2 modernisation
Second part in a series of commits to replace Gtk::OptionMenu widgets
with GParted::OptionComboBox.
This specific commit is about partition type combobox.
Closes!17 - Gtk2 modernisation
First part in a series of commits to replace Gtk::OptionMenu widgets
with GParted::OptionComboBox.
This specific commit is about partition alignment combobox.
Closes!17 - Gtk2 modernisation
The shades of aquamarine, cyan and orange didn't fit with the GNOME
32-colour palette. Create a set of aquamarine, cyan and orange shades
which match the GNOME palette shades and update the colours of the
relevant file systems accordingly.
Aquamarine Hilight (#97DFC7) -
Aquamarine Medium (#70D2B1) - NTFS
Aquamarine Dark (#3EA281) - REFS
Aquamarine Shadow (#1F7258) -
Cyan Hilight (#95E3E5) - EXTENDED
Cyan Medium (#6FCECE) -
Cyan Dark (#3C9899) -
Cyan Shadow (#166F70) -
Orange Hilight (#E59F6A) -
Orange Medium (#E58749) - BTRFS
Orange Dark (#C26825) - ZFS
Orange Shadow (#984F18) -
Note that the hues of aquamarine and cyan are quite close and for the
thin outlines of partitions used in GParted they aren't easy to
distinguish. Hence also using different lightness to additionally
separate the colour for extended partitions from NTFS and ReFS file
systems.
Just add detection of APFS using GParted's internal magic string
detection. It just matches 1 byte of the 2 byte object type and the
4 byte magic field found in the super block [1]. See code comment for
more details.
Blkid has just gained recognition of APFS with util-linux v2.33 released
06-Nov-2018 [2].
This will write enough for GParted's simple internal detection to find
APFS:
# python -c '
import sys
sys.stdout.write("\0"*24 + "\1\0" + "\0"*6 + "NXSB")
' > /dev/sdb1
[1] Apple File System Reference
https://developer.apple.com/support/apple-file-system/Apple-File-System-Reference.pdf
[2] [ANNOUNCE] util-linux v2.33
https://marc.info/?l=linux-fsdevel&m=154150400305928&w=2Closes#23 - GParted doesn't detect APFS (Apple File System)
Currently Linux Swap, Linux Suspend, and HFS use reds from the GNOME
32-colour palette with HFS Plus using serene red outside that palette.
HFSPLUS - Serene Red
HFS - Red Hilight
LINUX_SWAP - Red Medium
LINUX_SUSPEND - Red Dark
- Red Shadow
Apple have a new file system, APFS (Apple File System), which is a
successor to HFS Plus [1][2]. With HFS Plus using a colour outside the
GNOME 32-colour palette and there not being enough distinct reds
available for a new file system, create a new range of magenta colours
which fit with the GNOME palette and use them for the group of Apple
file systems.
Magenta Hilight (#D59FD4) - HFS
Magenta Medium (#B173B0) - HFSPLUS
Magenta Dark (#874986) - APFS
Magenta Shadow (#662C64) -
This commit just moves HFS and HFS Plus to their new magenta colours.
[1] About Apple File System
https://developer.apple.com/documentation/foundation/file_system/about_apple_file_system
"Overview
Apple File System replaces HFS Plus as the default file system for
iOS 10.3 and later, and for macOS High Sierra and later. Apple File
System offers improved file system fundamentals as well as several
new features, including cloning, snapshots, space sharing, fast
directory sizing, atomic safe-save, and sparse files.
"
[2] Apple File System Reference
https://developer.apple.com/support/apple-file-system/Apple-File-System-Reference.pdf
"About Apple File System
Apple File System is the default file format used on Apple
platforms. Apple File System is the successor to HFS Plus, so some
aspects of its design intentionally follow HFS Plus to enable data
migration from HFS Plus to Apple File System. Other aspects of its
design address limitations with HFS Plus and enable features such as
cloning files, snapshots, encryption, and sharing free space between
volumes.
"
Closes#23 - GParted doesn't detect APFS (Apple File system)
No other file system allows this, but btrfs allows simultaneous mounting
with different read-write permission. Further, btrfs allows resizing
via read-write mounts, but not via read-only mounts.
# mkfs.btrfs /dev/sdb1
btrfs-progs v4.15.1
...
Filesystem size: 512.00MiB
...
Number of devices: 1
Devices:
ID SIZE PATH
1 512.00MiB /dev/sdb1
# mount -o ro /dev/sdb1 /mnt/1
# mount -o rw /dev/sdb1 /mnt/2
# grep sdb1 /proc/mounts
/dev/sdb1 /mnt/1 btrfs ro,relatime,space_cache,subvolid=5,subvol=/ 0 0
/dev/sdb1 /mnt/2 btrfs rw,relatime,space_cache,subvolid=5,subvol=/ 0 0
# btrfs filesystem resize 1:500M /mnt/1
Resize '/mnt/1' of '1:500M'
ERROR: unable to resize '/mnt/1': Read-only file system
# echo $?
1
# btrfs file system resize 1:500M /mnt/2
Resize '/mnt/2' of '1:500M'
# echo $?
0
# btrfs filesystem show /dev/sdb1
Label: none uuid: 74ccd37a-e665-4f25-b77e-a305b8a025e9
Total devices 1 FS bytes used 128.00KiB
devid 1 size 500.00MiB used 88.00MiB path /dev/sdb1
Also with the above order of the read-only mount listed in /proc/mounts
first and the read-write mount second, GParted again allows a resize
operational to be tried, but if fails just like before:
Grow /dev/sdb1 from 512.00 MiB to 1.0 GiB (ERROR)
* calibrate /dev/sdb1 (SUCCESS)
* grow partition from 512.00 MiB to 1.00 GiB (SUCCESS)
* grow filesystem to fill the partition (ERROR)
* btrfs filesystem resize 1:max '/mnt/1' (ERROR)
Resize '/mnt/1 to '1:max'
ERROR: unable to resize '/mnt/1': Read-only file system
What happened is that the Mount_Info module only stores single read-only
flag against the mounted block device, not for each mount point, and as
the first and second sdb1 lines from /proc/mounts were processed, the
MountEntry became:
1st) mount_info[BS("/dev/sdb1")] -> {true , ["/mnt/1"]
2nd) mount_info[BS("/dev/sdb1")] -> {false, ["/mnt/1", "/mnt/2"]
So GParted thought the file system was mounted read-write, but used the
first mount point, /mnt/1, which was mounted read-only.
This is a very unusual situation so unlikely to be encountered by users.
Fix simply and safely by treating the mounted block device as mounted
read-only if any of the mount points are mounted read-only, rather than
just the last processed mount point.
Closes#10 - Gparted fails to resize btrfs partition that is mounted
read-only
Resizing a file system mounted read-only fails. Example:
# mkfs.btrfs /dev/sdb1
# mount -o ro /dev/sdb1 /mnt/1
In GParted try to resize partition sdb1. The operation fails like this:
Grow /dev/sdb1 from 512.00 MiB to 1.00 GiB (ERROR)
* calibrate /dev/sdb1 (SUCCESS)
* grow partition from 512.00 MiB to 1.00 GiB (SUCCESS)
* grow filesystem to fill the partition (ERROR)
* btrfs filesystem resize 1:max '/mnt/1' (ERROR)
Resize '/mnt/1' of '1:max'
ERROR: unable to resize '/mnt/1': Read-only file system
See GitLab issue for the testing results of attempting to online resize
all supporting file system while mounted read-only. No file system
allows online resizing while mounted read-only, except for reiserfs.
Issue #10 - Gparted fails to resize btrfs partition that is mounted
read-only
https://gitlab.gnome.org/GNOME/gparted/issues/10
Fix by preventing online resizing of *all* file systems mounted
read-only, including reiserfs. Instead of displaying the resize dialog
in this case, display an information dialog explaining why the partition
can't be resized. This is similar to what happens when attempting to
create a new partition on a disk without a partition table. The new
dialog looks like:
(!) Unable to resize read-only file system /dev/sdb1
The file system can not be resized while it is mounted read-only.
Either unmount the file system or remount it read-write.
[ OK ]
Closes#10 - Gparted fails to resize btrfs partition that is mounted
read-only
Set the partition read-only mount flag at the same time as setting the
file system mount points.
Closes#10 - Gparted fails to resize btrfs partition that is mounted
read-only
Parse file system mount options string from file and mount command
output, extracting the setting for the read-only flag and storing in the
mount maps. Read-only flag for swap space gets the struct MountEntry
constructor default of false.
Closes#10 - Gparted fails to resize btrfs partition that is mounted
read-only
Just updates the 2 maps in the Mount_Info module so that they also have
a read-only flag for each mount. Ensure that when a struct MountEntry
is created the readonly bool POD (Plain Old Data) type is initialised by
the constructor. Nothing yet sets or uses the flag.
Closes#10 - Gparted fails to resize btrfs partition that is mounted
read-only
Back when unallocated space handling was being added, this case was not
converted from open coding to using the provided method to check for
unknown file system usage. Specifically this commit missed using
Partition::sector_usage_known() in
Dialog_Base_Partition::prepare_new_partition():
7ebedc4bb3
Don't show intrinsic unallocated space (#499202)
Fix it now.
The function was using std::map::count() [1] to test if the file system
entry existed in the map before looking up the value using
std::map::operator[] to avoid having operator[] inserting elements which
don't exist [2].
Rewrite using std::map::find() [3] so that map is only searched once,
and so that it is more obvious what is happening without having to know
the subtleties of std::map::count() and ::operator[].
[1] std::map::count()
http://www.cplusplus.com/reference/map/map/count/
"Searches the container for elements with a key equivalent to k and
returns the number of matches.
Because all elements in a map container are unique, the function can
only return 1 (if the element is found) or zero (otherwise).
"
[2] std::map::operator[]
http://www.cplusplus.com/reference/map/map/operator[]/
"If k does not match the key of any element in the container, the
function inserts a new element with that key and returns a reference
to its mapped value. Notice that this always increases the
container size by one, even if no mapped value is assigned to the
element (the element is constructed using its default constructor).
"
[3] std::map::find
http://www.cplusplus.com/reference/map/map/find/
"Searches the container for an element with a key equivalent to k
and returns an iterator to it if found, otherwise it returns an
iterator to map::end.
"
There is no prospect of there being ufs-tools on Linux. The was a
project which did release ufs-tools version 0.1 in 2004, but has been
inactive since then.
http://ufs-linux.sourceforge.net/
Copying and moving is now implemented for file systems in the basic
supported category. Also mounting and unmounting of unsupported file
system and reporting their usage while mounted has been added. This is
all the support that GParted has ever implemented for UFS. Therefore
re-assign UFS as a basic supported file system as it looses no
functionality.
Closes!13 - Support copying and moving of unsupported partition content
For unsupported (including basic supported) file systems, also record
the mount point(s) when mounted and from /etc/fstab when not. This
allows mounted unsupported file systems to be unmounted and ones with
/etc/fstab entries to be mounted, just like fully supported file
systems.
Also for unsupported (again including basic supported) mounted file
systems query the kernel for the usage, just like is already done for
supported file systems.
Closes!13 - Support copying and moving of unsupported partition content
When previewing copying a partition of unknown file system usage into an
existing partition, the usage still shows that of the overwritten file
system. This affects existing supported file systems EXFAT, F2FS, MINIX
and UFS and the new basic supported one too, all for which GParted can't
read the file system usage.
Handle the case of the source file system usage being unknown and
explicitly set the copied usage to unknown too.
Closes!13 - Support copying and moving of unsupported partition content
GParted previews copying a partition of unknown file system usage into a
new partition as 100% used. This affects existing supported file
systems EXFAT, F2FS, MINIX and UFS and the new basic supported ones too,
all for which GParted can't read the file system usage.
When preparing the working new_partition object in the Copy / Paste
dialog, the maths for the known file system usage happened to convert
the figures of used = -1 and unused = -1 into set_sector_usage(-1, 0).
Those values passed to set_sector_usage() mean unable to query the file
system size so assume it fills the partition and unused is 0, hence 100%
used.
Fix this by specifically handling the copying of file systems with
unknown usage, setting the pasted file system usage to unknown too,
used = -1 and unused = -1.
Closes!13 - Support copying and moving of unsupported partition content
To display the supported actions for all basic supported file systems to
the users.
Prepare the list of file system actions in Win_GParted because calling
get_fs() for the "other" actions requires the gparted_core object and
load_filesystems() currently doesn't have access to it. One alternative
would have been to make get_fs() and FILESYSTEMS static members of
GParted_Core class. Another alternative would have been to pass the
gparted_core object to load_filesystems(). The chosen way seemed
simplest.
Closes!13 - Support copying and moving of unsupported partition content
Want a single term under which the supported actions for all basic
supported file systems are displayed in the File System Support dialog.
"Unknown" isn't the correct adjective because the group includes
unknown, but also includes: BitLocker, GRUB2 core image, ISO9660, Linux
SWRaid, Linux Suspend, REFS and ZFS. Add "other" file system type just
for displaying in the dialog.
Closes!13 - Support copying and moving of unsupported partition content
Introduce a third category of basic file system support to go along with
the existing full and none. Use the file system's entry in
FILESYSTEM_MAP to determine the level of support. See comment in
GParted_Core::init_filesystems() for details.
Add and remove FILESYSTEM_MAP NULL pointer entries as required, so that
only the file system types intended to have basic support have such
entries.
Closes!13 - Support copying and moving of unsupported partition content
PATCHSET OVERVIEW:
Forum user wanted to be able to move a partition with unknown content:
Topic: Can't move/rezise partition on android device (unknown format)
http://gparted-forum.surf4.info/viewtopic.php?id=17742
While GParted isn't going to be able to run any sort of file system
check on the unknown content there isn't any reason why such a partition
can't be copied or moved so long as the partition stays the same size.
GParted can just use it's existing internal block copy routine it uses
for copying and moving most partition content. This is no different to
a few of the already supported file system types which don't have a
check-repair tool: exfat, f2fs, nilfs2, udf, ufs.
This patchset introduces a third category called basic file system
support to go along with the existing full and unsupported categories.
Basic supported file systems will just use GParted's inbuilt
capabilities to perform actions so they won't need a derived FileSystem
implementation class. Unknown file systems along with all other
recognised, but otherwise unsupported, file systems will be assigned to
this new basic supported category.
THIS PATCH:
FS_UNKNOWN is used when GParted is unable to identify the contents of a
partition. FS_UNKNOWN is also used to generate a file system support
set with no supported actions, in the FileSystem::FS::FS() constructor
and in GParted_Core::get_fs().
As support for operations on partitions with unknown content is being
added, the second usage will be confusing or even wrong.
FS( FS_UNKNOWN ) constructs the no supported actions set, yet GParted
will support some actions for the FS_UNKNOWN file system type.
Therefore add FS_UNSUPPORTED for the second usage.
Closes!13 - Support copying and moving of unsupported partition content
No functional change. The code layout is old and a mess, not lining up
vertically. Use more common code layout and spaces to align text
vertically. List cases in enumeration order. Identify each colour
choice as either in the GNOME palette (no marking), an extended shade to
a colour in the GNOME palette [+], or a colour outside the GNOME palette
[*].
There's lots of other switch statements just in Utils.cc which could do
with tidying up, but this is the one I am looking at now.
FAT16 was a fully saturated green (RGB #00FF00) and FAT32 was a little
darker. These are out of character with the colours from the GNOME
palette for other file systems. Change the colours to use near
alternative Accent Greens from the GNOME colour palette. So now we have
the following file system colours, from light to dark:
FAT16 - Accent Green Hilight
FAT32 - Accent Green
EXFAT - Accent Green Dark
UDF - Accent Green Shadow
Strictly speaking only Accent Green and Accent Green Dark are part of
the GNOME palette. Accent Green Hilight and Accent Green Shadow are
extensions expanding the range of Accent Greens.
GNOME Human Interface Design 2.2.1 / Visual Design / colour /
https://developer.gnome.org/hig-book/2.32/design-color.html.en
"Guidelines
* Use the GNOME color palette. If you need a darker or lighter
shade, start from one of the colors from the palette and darken or
lighten as needed.
"
Util-linux package, at least as far back as version 2.23.2 as found on
CentOS 7, provides the mkfs.minix and fsck.minix commands. Also blkid
from the same package, recognises minix file systems.
Create version 3 file systems because MINIX 3 [1] is the only supported
version and that reportedly uses version 3 of the file system [2].
[1] MINIX 3 / History
https://en.wikipedia.org/wiki/MINIX_3#History
[2] Regarding MINIX 3 file system
https://groups.google.com/forum/#!topic/minix3/3-TeHR_23X8
"MINIX 3 uses Minix File System (MFS). More precisely MFS V3."
Closes!12 - Add minix file system support
I see the MINIX file system as a kind of forerunner to EXT* because of
it's history [1]. No body uses the original EXT file system any more,
however the MINIX file system is still used by the MINIX 3 operating
system. So use the same range of colours for MINIX and EXT2/3/4. Use
one shade darker blue for EXT2/3/4, allowing MINIX to use the lightest
blue. After adding MINIX support in the next patch, the colours will
become:
MINIX - Blue Hilight
EXT2 - Blue Medium
EXT3 - Blue Dark
EXT4 - Blue Shadow
[1] MINIX file system / History
https://en.wikipedia.org/wiki/MINIX_file_system#History
"When Linus Torvalds first started writing his Linux operating
system kernel (1991), he was working on a machine running MINIX, and
adopted its file system layout. This soon proved problematic, since
MINIX restricted filename lengths to fourteen characters (thirty in
later versions), it limited partitions to 64 megabytes, and the file
system was designed for teaching purposes, not performance. The
Extended file system (ext; April 1992) was developed to replace
MINIX's, but it was only with the second version of this, ext2, that
Linux obtained a commercial-grade file system. As of 1994, the
MINIX file system was "scarcely in use" among Linux users.
"
Closes!12 - Add minix file system support
If an EXT2/3/4 file system needs checking, then resize2fs will report an
error, rather than report the minimum file system size.
# mkfs.ext4 /dev/sdb11
# resize2fs -P /dev/sdb11
resize2fs 1.42.9 (28-Dec-2013)
Estimated minimum size of the filesystem: 17012
# debugfs -w -R "ssv state 0" /dev/sdb11
# resize2fs -P /dev/sdb11
resize2fs 1.42.9 (28-Dec-2013)
Please run 'e2fsck -f /dev/sdb11' first.
# echo $?
1
This will prevent GParted reading the file system usage and in turn
GParted won't allow the file system to be shrunk. Re-add the previous
method of reading the free space from dumpe2fs output as a fallback.
With this change, the worst case scenario is that GParted allows the
user to attempt to shrink an unclean EXT4 file system, smaller that that
which resize2fs allows and gets an error telling them so. As part of
the failed shrink operation GParted will have checked the file system so
on refresh GParted will get the correct minimum size next time.
This scenario only seems to apply to unclean EXT4 file systems because
resize2fs has a larger minimum size that the free blocks would suggest
because of extra space requirements when resizing EXT4 file systems [1].
[1] e2fsprogs 1.44.3, resize/resize2fs.c:calculate_minimum_resize_size()
https://git.kernel.org/pub/scm/fs/ext2/e2fsprogs.git/tree/resize/resize2fs.c?h=v1.44.3#n2946
/*
* For ext4 we need to allow for up to a flex_bg worth of
* inode tables of slack space so the resize operation can be
* guaranteed to finish.
*/
/*
* We need to reserve a few extra blocks if extents are
* enabled, in case we need to grow the extent tree. The more
* we shrink the file system, the more space we need.
*
* The absolute worst case is every single data block is in
* the part of the file system that needs to be evacuated,
* with each data block needs to be in its own extent, and
* with each inode needing at least one extent block.
*/
Closes#8 - Shrinking an EXT4 partition does not respect resize2fs
limits
A user reported GParted failed to shrink an EXT4 file system because
GParted tried to shrink it smaller than resize2fs reported minimum size.
Operation details were:
Shrink /dev/sdc1 from 931.51 GiB to 605.00 GiB (ERROR)
calibrate /dev/sdc1 (SUCCESS)
path: /dev/sdc1 (partition)
start: 63
end: 1953520064
size: 1953520002 (931.51 GiB)
check file system on /dev/sdc1 for errors and (if poss...(SUCCESS)
e2fsck -f -y -v -C 0 '/dev/sdc1' (SUCCESS)
...
158165624 blocks are used (64.77% of 244190000)
...
shrink file system (ERROR)
resize2fs -p '/dev/sdc1' 634389176K (ERROR)
resize2fs 1.44.2 (14-May-2018)
resize2fs: New size smaller than minimum (171882113)
The GParted figures:
* Partition size = 1953520064 (512b sectors) = 976760032 KiB
* FS size = 244190000 (4K blocks) = 976760000 KiB
* Used FS size = 158165624 (4K blocks) = 632662496 KiB
* Requested FS size = 634389176 KiB
The resize2fs figure:
* Minimum FS size = 171882113 (4K blocks) = 687528452 KiB
GParted uses the number of free blocks in the file system to determine
the minimum size it can shrink a file system to. However resize2fs uses
it's own internally calculated minimum size and won't shrink a file
system below that size, as seen in the above details. Resize2fs does
have a force flag, (-f) which overrides some safety checks which are
normally enforced, to allow it to try to shrink a file system smaller
than it's calculated minimum. GParted currently doesn't use the force
flag and it seems unwise for it to start to do so.
So for unmounted EXT2/3/4 file systems, change GParted to use
'resize2fs -P' to get the minimum file system size, rather than using
the number of free blocks direct from the super block, as reported by
'dumpe2fs -h'.
Mounted file systems still use statvfs() to provide file system usage.
As mounted EXT2/3/4 file systems can't be shrunk the fact that statvfs()
produces different, possibly smaller than minimum, figures than those
from 'resize2fs -P' doesn't matter.
Closes#8 - Shrinking an EXT4 partition does not respect resize2fs
limits
No functional change. Just work in FS block sized units until as late
as possible in ext2::set_used_sectors(), before converting to device
sector size units. This is to make the following change simpler and
easier to understand.
Closes#8 - Shrinking an EXT4 partition does not respect resize2fs
limits
This is functionally identical, but is just to follow established coding
pattern [1] of specifying the FSType when constructing struct FS, rather
and setting it afterwards. luks.cc was added after the aforementioned
commit, but was being developed in parallel so was created [2] following
the old coding pattern.
[1] 1a4cefb960
Initialise all struct FS members
[2] 070d734e57
Add busy detection of LUKS mapping (#760080)
Shrinking an LVM2 Physical Volume on CentOS 7 with the latest
lvm2 2.02.177 fails like this:
Shrink /dev/sda9 from 1.00 GiB to 768.00 MiB
* calibrate /dev/sda9
* check file system on /dev/sda9 for errors and (if possib...(SUCCESS)
* shrink file system (ERROR)
* lvm pvresize -v --setphysicalvolumesize 786432K '/dev/...(ERROR)
0 physical volume(s) resized / 1 physical volume(s) not resized
Wiping internal VG cache
Wiping cache of LVM-capable devices
/dev/sda9: Requested size 712.00 MiB is less than real size 1.00 GiB. Proceed? [y/n]:[n]
Physical Volume /dev/sda9 not resized.
This upstream change to lvm2 [1] makes pvresize prompt for confirmation
whenever the --setphysicalvolumesize option is used. (The change was
included in lvm2 2.02.171 and later, which is used in recent
distributions. The reporter found the issue on Ubuntu 18.04 LTS and I
reproduced the issue on RHEL/CentOS 7.5). The set size option has to be
used when shrinking the PV before shrinking the partition therefore fix
this issue by adding lvm common option --yes when using the set size
option.
[1] https://sourceware.org/git/?p=lvm2.git;a=commitdiff;h=cbc69f8c693edf0d1307c9447e2e66d07a04bfe9
pvresize: Prompt when non-default size supplied.
Closes#1 - Can't shrink LVM partition due to pvresize prompt
After a failed LUKS unlock attempt the password entry dialog shows the
error "Failed to open LUKS encryption". Improve the user experience by
clearing that error message at the start of the next attempt to avoid
contradictory information with the main windows status of "Opening
encryption on $PARTITION" whilst performing the next unlock attempt.
Bug 795617 - Implement opening and closing of LUKS mappings
When the wrong LUKS password is entered and the [Unlock] button clicked,
the wrong password is left in the entry box and focus remains on the
[Unlocked] button. Improve the user experience by selecting
(highlighting) the whole of the wrong password ready for deletion or
retyping and ensuring that the entry box always has focus.
Just for completeness also programmatically make the password entry box
have focus when the dialog box is created and first displayed, even
though it gets this by default.
Bug 795617 - Implement opening and closing of LUKS mappings
We previously migrated our web site from http://gparted.org to
https://gparted.org under:
bug 786707 - gparted.org does not use HTTPS
and updated URLs in the GParted Manual to match in commit:
a8172ecb04
Convert Manual links to HTTPS where possible and update version
Now update the URLs displayed in the GParted application too.
Bug 796411 - Enhancements request - URL links
The Change_UUID_Warning vectors were fat16 and ntfs class member
variables, but are only ever accessed in the get_custom_text() method.
Make them local variables in get_custom_text() instead. Static so that
references to them can be returned.
I completely missed that when reading XFS file system size and usage it
was using the '-r' read-only flag to xfs_db because it was at the end of
the string on the following line of code. Move it to the start of the
xfs_db command line, like it is when reading the file system label.
Same case as for FileSystem Label dialog before; the Partition Name
dialog only has a single line of just 2 widgets. Therefore switch to a
simpler horizontal box widget to lay them out.
The FileSystem Label dialog only has a single line of just 2 widgets; a
text label and entry box widget. There is no need to use a multi-line
capable table to hold this. Switch to a simpler horizontal box widget.
Note that this change is not related to porting to Gtk 3 and stopping
using deprecated APIs because both HBox [1] and Table [2] are deprecated
in Gtk 3.2 and Gtk 3.4 and replaced by Box with horizontal orientation
and Grid respectively.
[1] NEWS file from gtkmm 3.2, actually first released in gtkmm 3.1.6
(unstable):
https://git.gnome.org/browse/gtkmm/tree/NEWS?h=3.2.0#n91
"Gtk:
* All H* or V* specialized classes have been deprecated, to
match the deprecations in the GTK+ C API. You should now
set the orientation instead.
This includes HBox, VBox, HButtonBox, VButtonBox, HPaned,
VPaned, HScale, VScale, HSeparator, VSeparator, HScrollbar
and VScrollbar."
[2] NEWS file from gtkmm 3.4, actually first released in gtkmm 3.3.2
(unstable):
https://git.gnome.org/browse/gtkmm/tree/NEWS?h=3.4.0#n162
"* Deprecate Gtk::Table in favour of Gtk::Grid."
Replace return by value of const strings from
FileSystem::get_custom_text() and get_generic_text() because that
implies duplication of those strings. Return a reference to constant
strings instead.
Compiling (with new enough g++) produces this warning:
PasswordRAMStore.cc: In member function 'void GParted::PWStore::erase_all()':
PasswordRAMStore.cc:177:2: warning: this 'if' clause does not guard... [-Wmisleading-indentation]
if ( protected_mem != NULL );
^~
PasswordRAMStore.cc:193:3: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'if'
memset( protected_mem, '\0', ProtectedMemSize );
^~~~~~
Looks like a stray semicolon...
Bug 796293 - Fix null pointer check accidentally disabled
Replace the insert() method (which reports an error when inserting a
password with a key which already exists) with the store() method which
replaces or inserts the password depending on whether the key already
exists or not respectively. There is also an optimisation that nothing
is changed if the password to be replaced is the same as the one already
stored. The code in Win_GParted::open_encrypted_partition() is
simplified now it doesn't have to implement this pattern of behaviour
itself.
Bug 795617 - Implement opening and closing of LUKS mappings
Reports generic GParted error "Failed to open LUKS encryption" on any
failure unlocking the partition. Choosing not to display cryptsetup
reported errors because those messages and their translations are not
under GParted control.
Bug 795617 - Implement opening and closing of LUKS mappings
The underlying C coded Gtk Entry widget is careful to zero memory after
use, allowing the widget to be safely used for password entry [1].
However the C++ method Gtk::Entry::get_text() just takes the underlying
C string from the Gtk Entry widget and copies it when constructing a
Glib::ustring for the return value [2].
So directly use the Gtk/C API to get the C string instead.
[1] https://git.gnome.org/browse/gtk+/tree/gtk/gtkentrybuffer.c?h=3.22.28#n92
See function trash_area() which zeros memory and its use in
gtk_entry_buffer_normal_insert_text(),
gtk_entry_buffer_normal_delete_text() and
gtk_entry_buffer_finalize().
[2] https://git.gnome.org/browse/gtkmm/tree/gtk/src/entry.hg?h=3.22.2#n104
_WRAP_METHOD(Glib::ustring get_text() const, gtk_entry_get_text)
https://git.gnome.org/browse/glibmm/tree/docs/internal/using_gmmproc.txt?h=2.46.1#n53
_WRAP_METHOD(Glib::ustring METHOD const, FUNC) is processed to:
Glib::ustring METHOD() const
{
return Glib::convert_const_gchar_ptr_to_ustring(
FUNC(const_cast<GtkEntry*>(gobj())));
}
https://git.gnome.org/browse/glibmm/tree/glib/glibmm/utility.h?h=2.46.1#n82
Glib::ustring convert_const_gchar_ptr_to_ustring(const char* str)
{
return (str) ? Glib::ustring(str) : Glib::ustring();
}
So Gtk::Entry::get_text() calls Glib::ustring() constructor which
copies the C string to create the Glib::ustring object returned.
Bug 795617 - Implement opening and closing of LUKS mappings
To keep password dialog open, just keep running it in a loop performing
LUKS mapping unlock attempts with the entered passphrase until it
succeeds or the dialog is cancelled or closed. This is the same model
that is already used for the File Support System dialog and how the
[Rescan For Supported Actions] button is implemented.
Also any error from attempting to open the LUKS mapping is no longer
displayed in a separate error dialog or at all. Will add some sort of
error reporting into the password entry dialog in a following commit.
Creates new method Win_GParted::open_encrypted_partition() which handles
the non UI parts of attempting to open an encrypted partition. Running
"cryptsetup luksOpen" and updating the stored passphrase as needed.
Bug 795617 - Implement opening and closing of LUKS mappings
Initial addition of a password entry dialog. Looks like:
+------------------------------------------------+
| LUKS Passphrase /dev/sdb1 |
+------------------------------------------------+
| Enter LUKS passphrase to open /dev/sdb1 |
| Passphrase: [ ] |
| |
| [ Cancel ] [ Unlock ] |
+------------------------------------------------+
A standard Gtk Dialog is used to accept the password once, with any
errors displayed in a separate error dialog afterwards. This is poor UI
design. A password dialog should remain open for all authentication
attempts and only close when successful or the dialog is cancelled or
closed. This UI design issue will be improved in following commits.
Bug 795617 - Implement opening and closing of LUKS mappings
Implement Close Encryption partition menu item.
The Open Encryption action is not yet implemented and instead reports an
error detailing the open encryption command. A dialog needs to be
written to accept the password entry and pass it to the open encryption
command.
Bug 795617 - Implement opening and closing of LUKS mappings
Add new item to the partition menu to allow the user to open and close
the LUKS mapping. However for now the menu item is always disabled and
there is no implementation behind it to actually open or close the LUKS
mapping. Fragment of the partition menu is now:
...
Format to >
-----------------
Open Encryption <- New menu item
Mount
-----------------
Name Partition
...
Has to be two separate menu items to clearly represent to the user that
LUKS mappings and file system mounting are two separate busy states.
And also in the case of an open but unmounted file system to offer both
actions; close encryption and mount file system.
The text of the menu item automatically changes similarly to how it does
for the Mount/Unmount, Swapon/Swapoff, Activate/Deactivate item
depending on the state of the LUKS mapping. For open LUKS mappings it
will show "Close Encryption" and for all other cases (closed LUKS
mapping or partition is not encrypted) "Open Encryption". Again similar
to how the default of "Mount" is shown for unallocated and unknown
partitions.
Bug 795617 - Implement opening and closing of LUKS mappings
In preparation for adding the ability to toggle the encryption busy
state (open/close the encryption volume), rename existing members to
reflect that they are related to changing the file system state. (Swap
and LVM2 Physical Volumes are handled as file systems by GParted).
class Win_GParted renaming:
MENU_TOGGLE_BUSY -> MENU_TOGGLE_FS_BUSY
allow_toggle_busy_state() -> allow_toggle_fs_busy_state()
toggle_busy_state() -> toggle_fs_busy_state()
check_toggle_busy_allowed() -> check_toggle_fs_busy_allowed()
Bug 795617 - Implement opening and closing of LUKS mappings
Replace echoing "dm" into jfs_debugfs via a shell command to directly
writing "dm" to the input of the jfs_debug command. One less use of the
shell.
Bug 795617 - Implement opening and closing of LUKS mappings
As discussed in "LUKS password handling, threats and preventative
measures" [1] GParted must be able to pass LUKS passphrases to
cryptsetup via standard input to avoid having to write passwords to the
file system and deal with additional security requirements. Therefore
add a way to write input into created child processes. For small
amounts of input, writing up to the pipe buffer capacity won't block
[2]. This is 64K on versions of Linux in any currently supported
distributions.
[1] LUKS password handling, threats and preventative measures
https://bugzilla.gnome.org/show_bug.cgi?id=627701#c56
GParted must not become a password manage so it must never save
LUKS passwords to disk across separate invocations of GParted.
...
GParted should avoid writing a temporary file containing the LUKS
password as it introduces extra complexity with trying to safely
handle and erase file content. Instead GParted must
programmatically pass the LUKS password via standard input to the
cryptsetup command.
[2] pipe(7) manual page:
Pipe capacity
A pipe has a limited capacity. If the pipe is full, then a
write(2) will block or fail, depending on whether the O_NONBLOCK
flag is set (see below). ...
In Linux versions before 2.6.11, the capacity of a pipe was the
same as the system page size (e.g., 4096 bytes on i386). Since
Linux 2.6.11, the pipe capacity is 65536 bytes.
Bug 795617 - Implement opening and closing of LUKS mappings
Use private access into the PasswordRAMStore class to directly obtain
the address of the locked memory, rather than inferring it from the
address of the first stored password. This simplifies
PasswordRAMStoreTest::SetUpTestCase() and avoids encoding most of the
implementation knowledge that the first password will be stored at the
start of the protected memory.
Bug 795617 - Implement opening and closing of LUKS mappings
Test that all passwords are zeroed by PasswordRAMStore::erase_all(), the
same method as used in the PasswordRAMStore destructor.
Bug 795617 - Implement opening and closing of LUKS mappings
Move zeroing of the locked memory into separate PWStore::erase_all()
private method. Then use this in the PWStore destructor. This is so
that zeroing of all passwords can be unit tested independently of
destructing the singleton PWStore object.
Bug 795617 - Implement opening and closing of LUKS mappings
Application level requirements for secure password management were set
out in "LUKS password handling, threats and preventative measures" [1].
The requirements are:
1) Passwords are stored in RAM and are not allowed to be paged to swap.
(However hibernating with GParted still running will write all of RAM
to swap).
2) Passwords are wiped from RAM when no longer needed. When each
password is no longer needed and when GParted closes.
3) Passwords are referenced by unique key. Recommend using LUKS UUIDs
as the unique key.
(Each LUKS password should only ever need to be entered once for each
execution of GParted. Therefore the passwords can't be stored in any
of the existing data structures such as Partitions or LUKS_Info cache
because all of these are cleared and reloaded on each device
refresh).
There seems to be two possible implementation methods: use an existing
library to provide secure memory handling, or write our own.
Libgcrypt [2] and libsodium [3] cryptographic libraries both provide
secure memory handling. (Secure memory is quite simple really, some
virtual memory locked into RAM which is zeroed when no longer needed).
Linking to an encryption library just to provide secure memory seems
like using a sledge hammer to crack a nut. Also because of requirement
(3) above a module is needed to "own" the pointers to the passwords in
the secure memory. Managing the secure memory ourselves is probably no
more code that that needed to interface to libgcrypt. Therefore handle
the secure memory ourselves.
So far the module is only compiled. It is not used anywhere in GParted.
[1] LUKS password handling, threats and preventative measures
https://bugzilla.gnome.org/show_bug.cgi?id=627701#c56
[2] libgcrypt general purpose cryptographic library, as used in GNU
Privacy Guard
https://gnupg.org/related_software/libgcrypt/
[3] libsodium crypto library
https://download.libsodium.org/doc/
Bug 795617 - Implement opening and closing of LUKS mappings
Increase the minimum required version of gtkmm to 2.16.0, thus allowing
removal of HAVE_GTK_SHOW_URI autoconf definition and associated fallback
code.
Bug 794253 - Desupport RHEL / CentOS 5 and raise minimum required
versions to glibmm 2.14.0 and gtkmm 2.16.0
Increase the minimum required version of gtkmm to 2.11.1, thus allowing
removal of:
* HAVE_SET_DEFAULT_ICON_NAME autoconf definition and associated
optional code.
* INSTALL_PIXMAPS_DIR automake conditional and associated make
instructions.
This is reversing these 3 commits, except for the higher minimum gtkmm
version:
1) a042107883
Only use Gtk::Window::set_default_icon_name method when available (#695279)
2) b09d6035cd
Add fallback method for specifying GParted icon (#695279)
3) d6baac2546
Only install fallback icon when required (#695279)
Bug 794253 - Desupport RHEL / CentOS 5 and raise minimum required
versions to glibmm 2.14.0 and gtkmm 2.16.0
Increase the minimum required version of glibmm to 2.14.0, thus allowing
removal of the HAVE_GLIB_REGEX autoconf definition and associated
conditional code. This is reversing commit, except for the new glibmm
minimum check:
456932846b
Implement fallback if Glib::Regex class is missing (#695279)
Bug 794253 - Desupport RHEL / CentOS 5 and raise minimum required
versions to glibmm 2.14.0 and gtkmm 2.16.0
E2fsprogs 1.41.0 (from 10 July 2008) first included ext4 support [1].
As RHEL / CentOS 6 is now the oldest supported distribution, and that
includes e2fsprogs 1.41.12 (from 22 August 2009) [2] all the e2fs
programs support ext4 so it is no longer necessary to also depend on
finding mkfs.ext4 before enabling each supported capability for ext4.
This makes the ext2::get_filesystem_support() look like all the others
in which each supported capability only depends on the presence of the
relevant file system specific command.
[1] Release notes for the e2fsprogs package / E2fsprogs 1.41.0
http://e2fsprogs.sourceforge.net/e2fsprogs-release.html#1.41.0
[2] pkgs.org > CentOS 6 > CentOS x86_64 > e2fsprogs
https://centos.pkgs.org/6/centos-x86_64/e2fsprogs-1.41.12-23.el6.x86_64.rpm.html
Bug 794253 - Desupport RHEL / CentOS 5 and raise minimum required
versions to glibmm 2.14.0 and gtkmm 2.16.0
PATCHSET OVERVIEW:
As of 31 March 2017 RHEL / CentOS 5 reached the end of their support
[1][2]. Therefore remove code which supports them. This makes RHEL /
CentOS 6 the oldest supported distribution. So the minimum required
versions of glibmm and gtkmm can be increased dropping some autoconf
checks and conditional code supporting older versions of these
libraries. This will undo the bulk of these these previous bug fixes:
* GParted 0.21.0
Bug 738706 - Add support for ext4 on RHEL/CentOS 5.x
* GParted 0.16.1
Bug 695279 - Fix GParted doesn't compile on RHEL / CentOS 5.9
[1] Red Hat Enterprise Linux Life Cycle
https://access.redhat.com/support/policy/updates/errata#Life_Cycle_Dates
[2] Subject: CentOS Linux 5 EOL
https://lists.centos.org/pipermail/centos-announce/2017-April/022350.html
THIS PATCH:
Remove checks for e4fsprogs commands, removing support for ext4 on
RHEL / CentOS 5.x. This is reverting earlier commit:
f672f68863
Check for e4fsprogs commands for ext4 support on RHEL/CentOS 5.x (#738706)
Mkfs_cmd member variable is being kept as a convenience so that it is
created once rather than on each use. Also note that as it is a
Glib::ustring type object, it's constructor will be called which will
initialise it to the empty string so it doesn't need initialising to the
empty string in the initialiser list of the ext2() constructor itself.
Bug 794253 - Desupport RHEL / CentOS 5 and raise minimum required
versions to glibmm 2.14.0 and gtkmm 2.16.0
Struct FS and struct FS_Limits are strongly related to the FileSystem
class, both being return values from members and associated with storing
file system attributes. Move their definitions from Utils.h into
FileSystem.h.
There are too many different types of things named "filesystem" in the
GParted code with the potential to cause confusion. Namely:
std::vector<FS> FILESYSTEMS
Vector of file system capabilities.
class FileSystem Base class interfacing to file system
specific executables for querying and
modification.
enum FILESYSTEM Symbolic constants representing each file
system type.
Many recent written or re-written functions already used a variable
named fstype. Rename enum FILESYSTEM to enum FSType to clearly
distinguish it from the other things with very similar names. Only
changing the name of the enumeration, not the name of variables of that
type too because that is a lot more lines of code and those can be
changed when the relevant code is re-written.
Try to format an existing partition with a file system which doesn't
fit. The error dialog reporting the partition as too small or too large
always claimed the file system was encrypted, whether it was or not.
For example trying to format a 128 MiB partition as btrfs produces this
error dialog:
(-) Cannot format this file system to [Encrypted] btrfs
A [Encrypted] btrfs file system requires
a partition of at least 256.00 MiB.
[ OK ]
This commit:
88136c96d7
Extend functions generating encrypted file system string (#774818)
just completely missed handling the case for non-encrypted file systems
in Utils::get_filesystem_string(). Add the missed code.
There are multiple repetitions of the same code getting a FileSystem
object, checking for NULL and then calling the file system specific
get_filesystem_limits(). Extract that into a common function.
GParted_Core::get_filesystem_limits() can't use the file system from the
passed Partition object because that is the current file system which
will be different from the intended file system for new and format
operations. So would look up the wrong derived FileSystem specific
object and call the wrong get_filesystem_limits(). Hence still needing
fstype as a separate parameter to pass the intended file system.
Bug 787204 - Minimum and maximum size of the UDF partition/disk
UDF file system minimum and maximum size limits are defined in terms of
numbers of file system blocks. So when resizing an existing file system
compute the byte size limits from the existing UDF file system's block
size. Alternatively when creating a new UDF file system use the
device's sector size as the multiplier instead.
Bug 787204 - Minimum and maximum size of the UDF partition/disk
As described in the previous commit, this is so that file system
specific implementations can dynamically determine size limits based on
Partition object attributes: such as the device sector size and the file
system block size. (Assuming set_used_sectors() sets
partition.fs_block_size for the type of file system in question).
Bug 787204 - Minimum and maximum size of the UDF partition/disk
Background information about UDF is that when creating a file system
it's block size must match the underlying device's sector size. For
optical media like CDs and DVDs that is 2K. For hard drives that is
usually 512 bytes or 4K. However if a UDF file system has been copied
from a device with a different sector size the UDF block size won't
match the sector size. Linux will happily mount such UDF file system.
Therefore the derived udf::get_filesystem_limits() will need access to
the file system block size when determining the size limits of an
existing UDF file system being resized and use the device sector size
when a new UDF file system is being created. All this can be queried
from an appropriate Partition object passed to get_filesystem_limits().
All the calls to get_filesystem_limits() have an appropriate Partition
object available already, except in Win_GParted::activate_reformat()
when composing a format operation. Or more correctly
activate_reformat() constructs temp_ptn, a suitable Partition object,
including with fs_block_size member defaulting to -1 indicating not a
resize, but not until after the file system size limits had been checked
and get_filesystem_limits() called.
Therefore reorder the code in activate_paste() so that the file system
size limits are checked after the wanted Partition object has been
created. No functional change with this commit.
Bug 787204 - Minimum and maximum size of the UDF partition/disk
All the code has been switched to call get_filesystem_limits() and use
struct FS_Limits. Remove struct FS members .MIN & .MAX.
Bug 787204 - Minimum and maximum size of the UDF partition/disk
Change Dialog_Partition_New to use a fs_limits rather than struct FS
and .MIN and .MAX. No passing of struct FS_Limits required. Just use
the FILESYSTEMS vector of struct FS to provide the file system type and
look up it's size limits each time the selection changes.
Bug 787204 - Minimum and maximum size of the UDF partition/disk
Refactor Win_GParted::activate_resize() to query the file system size
limits using the new get_filesystem_limits() method and pass those
limits into the dialog class as struct FS_Limits.
Bug 787204 - Minimum and maximum size of the UDF partition/disk
Mike Fleetwood