Quoting the relevant comments from GParted_Core::calibrate_partition():
Re-add the real partition path from libparted.
When creating a copy operation by pasting into unallocated space the
list of paths for the partition object was set to
["Copy of /dev/SRC"] because the partition didn't yet exist before
the operations were applied. Additional operations on that new
partition also got the list of paths set to ["Copy of /dev/SRC"].
This re-adds the real path to the start of the list making it
["/dev/NEW", "Copy of /dev/SRC"]. This provides the real path for
file system specific tools used during those additional operations
such mkfs for the format operation or fsck and others for the
resize/move operation.
FIXME: Having this work just because "/dev/NEW" happens to sort
before "Copy of /dev/SRC" is ugly! Probably have a separate display
path which can be changed at will without affecting the list of real
paths for the partition.
Having a separate display path is overly complicated and unnecessary.
Just replace the list of paths with the real one reported by libparted
if it contained "Copy of /dev/SRC", determined by checking if the file
exists. Otherwise continue to add the libparted name as an alternate
path. Whole disk devices can never be named "Copy of /dev/SRC" because
they are not partitioned so never created or deleted by GParted, only
ever written to, hence don't need the extra exists test logic.
Bug 766349 - Resolve code ugliness with partition path getting set to
"copy of /dev/SRC"
When composing a copy operation it always named the destination
partition as "copy of /dev/SRC". For the case of pasting into
unallocated space creating a new partition this was the right thing to
do as the partition doesn't yet exist so the path is not yet known.
However for the case of pasting into an existing partition the path is
known and replacing it with "copy of /dev/SRC" is wrong. No other
operation when operating on an existing partition changes it path.
Given a set of existing partitions, sdb1 to sdb4, compose a set of copy
operations as: copy sdb1 to sdb2, copy sdb2 to sdb3 and copy sdb3 to
sdb4. The displayed partitions before being applied become:
/dev/sdb1
copy of /dev/sdb1
copy of copy of /dev/sdb1
copy of copy of copy of /dev/sdb1
And the pending operations are named:
Copy /dev/sdb1 to /dev/sdb2
Copy copy of /dev/sdb1 to /dev/sdb3
Copy copy of copy of /dev/sdb1 to /sev/sdb4
This is perverse. In the case of pasting into an existing partition
keep the real path name. This keeps the partitions being displayed as:
/dev/sdb1
/dev/sdb2
/dev/sdb3
/dev/sdb4
And the pending operations named as:
Copy /dev/sdb1 to /dev/sdb2
Copy /dev/sdb2 to /dev/sdb3
Copy /dev/sdb3 to /dev/sdb4
Much more understandable.
Also switch to an upper case "C" in "Copy of /dev/SRC" as the temporary
path name when pasting into unallocated space. Finally update the
comment in calibrate_partition() to describe the remaining cases when
re-adding the path is still required.
Bug 766349 - Resolve code ugliness with partition path getting set to
"copy of /dev/SRC"
Make the code a little more self documenting by adding the symbolic
constants:
SETTLE_DEVICE_APPLY_MAX_WAIT_SECONDS
SETTLE_DEVICE_PROBE_MAX_WAIT_SECONDS
which highlight that settle_device() is called in two different
contexts, device probe and apply operations, with two different timeout
values.
File system specific commands sometimes fail reporting that the
partition specific /dev entry doesn't exist. Example failing check
operation details:
Check and repair file system (ext4) on dev/sdb4
calibrate /dev/sdb4
path: /dev/sdb4 (partition)
start: 4196352
end: 6293503
size: 2097152 (1.00 GiB)
check file system on /dev/sdb4 for errors and (if possible) fix them
e2fsck -f -y -bv -C 0 /dev/sdb4
e2fsck 1.42.9 (28-Dec-2013)
e2fsck: No such file or directory while trying to open /dev/sdb4
Possibly non-existent device?
This has been reproduced on CentOS 7. Debugging shows that the
libparted calls used to re-read the partition details in
GParted_Core::calibrate_partition() leads to udev removing and re-adding
all the partition /dev entries for the disk.
# udevadm monitor &
# gpartedbin
...
16.480662 +12.618659 calibrate_partition() calling get_device("/dev/sdb", lp_device) ...
16.483644 +0.002982 calibrate_partition() get_device() returned
16.483678 +0.000034 calibrate_partition() calling get_disk(lp_device, lp_disk) ...
16.618113 +0.134435 calibrate_partition() get_disk() returned
KERNEL[19275.707968] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb1 (block)
16.618561 +0.000448 destroy_device_and_disk() calling ped_disk_destroy(lp_disk) ...
16.618584 +0.000023 destroy_device_and_disk() ped_disk_destroy() returned
16.618591 +0.000007 destroy_device_and_disk() calling ped_device_destroy(lp_disk) ...
16.618602 +0.000011 destroy_device_and_disk() ped_device_destroy() returned
16.618687 +0.000085 calibrate_partition() return true
16.618851 +0.000164 execute_command() e2fsck -f -y -v -C 0 /dev/sdb4
KERNEL[19275.708389] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb2 (block)
KERNEL[19275.708500] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb3 (block)
KERNEL[19275.708643] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb4 (block)
KERNEL[19275.768278] change /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb (block)
KERNEL[19275.771171] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb1 (block)
KERNEL[19275.771360] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb2 (block)
KERNEL[19275.771542] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb3 (block)
KERNEL[19275.775858] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb4 (block)
UDEV [19275.820153] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb3 (block)
UDEV [19275.823152] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb4 (block)
UDEV [19275.828275] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb1 (block)
16.742735 +0.123884 execute_command() exit status 8
UDEV [19275.841425] remove /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb2 (block)
UDEV [19275.905478] change /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb (block)
UDEV [19276.013580] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb3 (block)
UDEV [19276.034728] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb4 (block)
UDEV [19276.174840] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb1 (block)
UDEV [19276.237105] add /devices/pci0000:00/0000:00:0d.0/ata4/host3/target3:0:0/3:0:0:0/block/sdb/sdb2 (block)
So exactly when GParted is running the external e2fsck command, udev is
in the middle of removing and re-adding all the /dev partition entries
for the disk. Hence the above failure reporting that /dev/sdb4 didn't
exist. This error depends on the timing between GParted running the
external file system specific command and udev removing and re-adding
the entries, so sometimes it works and sometimes it fails.
Further debugging showed that simply opening and closing the whole disk
device read-write triggers the same removing and re-adding of all the
partition /dev entries with udev >= 219. Opening the whole disk device
read-write is what libparted has always done until this post
libparted 3.2 patch to make it open read-only when probing:
http://git.savannah.gnu.org/cgit/parted.git/commit/?id=44d5ae0115c4ecfe3158748309e9912c5aede92d
libparted: Use read only when probing devices on linux (#1245144)
To fix this simply wait for udev devices to settle in
calibrate_partitions(). The longest I have seen udev take to do this is
0.80 seconds in a VM. Wait up to 10 seconds as is done in commit() ->
commit_to_os(), also called when applying operations.
On configurations which don't have this issue execution of udevadm
settle, which will return immediately, adds at most 0.1 seconds to the
time taken for the calibrate step. This won't be noticed in the time
taken of the operation details so there is no point in trying to avoid
executing udevadm settle when not needed.
Bug 762941 - Operations sometimes failing with: No such file or
directory
Minor issues:
1) In the while loop reading from /proc/partitions into variable line,
just after the sscanf() call the variable was re-purposed to hold the
device name making the code unnecessarily hard to follow.
2) Variable c_str was a fixed sized buffer holding the device name read
from /proc/partitions.
3) Variable c_str name provides no meaning as to what data it holds.
4) Return value from all the Utils::regexp_label() calls is converted
from Glib::ustring to std::string to be stored in device variable.
Resolve by using Utils::regexp_label() to extract the device name from
each line in /proc/partitions and store in the variable device, already
used for this purpose and now changed to type Glib::ustring.
realpath(3) manual page says:
BUGS
The POSIX.1-2001 standard version of this function is broken by
design, since it is impossible to determine a suitable size for
the output buffer, resolved_path. According to POSIX.1-2001 a
buffer of size PATH_MAX suffices, but PATH_MAX need not be a
defined constant, and may have to be obtained using pathconf(3).
And asking pathconf(3) does not really help, since, on the one
hand POSIX warns that the result of pathconf(3) may be huge and
unsuitable for mallocing memory, and on the other hand
pathconf(3) may return -1 to signify that PATH_MAX is not
bounded. The resolved_path == NULL feature, not standardized in
POSIX.1-2001, but standardized in POSIX.1-2008, allows this
design problem to be avoided.
The resolved_path == NULL feature of realpath() has existed as a Glibc
extension since realpath() was first added to Glibc 1.90, released in
June 1996. Therefore it can be used unconditionally.
https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=fa0bc87c32d02cd81ec4d0ae00e0d943c683e6e1
Bug 764369 - Use realpath() safely
The previous commit (Fix crash reading NTFS usage when there is no
/dev/PTN entry) identified that the FileSystem member variable "index"
is too small on 64-bit machines. Also this member variable stores no
FileSystem class information and was being used as a local variable.
Replace with local variables of the of the correct type, wide enough to
store the npos not found value.
Bug 764658 - GParted crashes when reading NTFS usage when there is no
/dev/PTN entry
On a 64-bit distribution, with an NTFS file system in a partition
without a /dev entry then GParted will crash when attempting to read
the file system usage. Not having a /dev entry for the partition is
rare and only known to occur for the disk devices used within Fake RAID
(dmraid) arrays, and then only on Ubuntu 12.04 LTS. Other/newer
distributions do create /dev entries for partitions found on disk
devices within Fake RAID arrays.
Create mirror Fake RAID array:
# dmraid -f isw -C MyArray --type 1 --disk /dev/sdc,/dev/sdd
# dmraid -ay
Create NTFS partition on the Fake RAID array. On refresh GParted
crashes:
# ./gpartedbin
(gpartedbin:590): glibmm-ERROR **:
unhandled exception (type std::exception) in signal handler:
what: basic_string::assign
Without a /dev/sdc1 device entry the ntfsresize command reports this:
# ntfsresize --info --force --no-progress-bar /dev/sdc1
ntfsresize v2015.3.14 (libntfs-3g)
ERROR(2): Failed to check '/dev/sdc1' mount state: No such file or directory
Probably /etc/mtab is missing. It's too risky to continue. You might try
an another Linux distro.
The problem code in ntfs::set_used_sectors():
145 index = output.find( "Cluster size" );
146 if ( index == output.npos ||
147 sscanf( output.substr( index ).c_str(), "Cluster size : %Ld", &S ) != 1 )
As "Cluster size" did not exist in the output find() returned the not
found token of string::npos [1], which in a 64-bit environment is
represented by 2^64-1 [2]. However it was saved in the variable index
of type unsigned integer, which is only a 32-bit integer, thus
truncating it to 2^32-1. Therefore the comparison failed and sscanf()
tried to parse the output starting at offset 2^32-1 which resulted in
the crash.
Introduced by commit:
324d99a172
Record file system block size where known (#760709)
Fix by following the same pattern of the other comparisons in
ntfs::set_used_sectors() which checks if index is less than the output
length.
References:
[1] std::string::find
http://www.cplusplus.com/reference/string/string/find/
[2] std::string::npos
http://www.cplusplus.com/reference/string/string/npos/
(Note that Glib::ustring is derived from std::string in the Standard C++
library and provides a compatible interface).
Bug 764658 - GParted crashes when reading NTFS usage when there is no
/dev/PTN entry
GParted is allowing creation of a FAT32 formatted partition of any size.
However with a 512 byte sector size the maximum volume size of a FAT32
file system is reported to be 2 TiB.
* Wikipedia: File Allocation Table / FAT32
https://en.wikipedia.org/wiki/File_Allocation_Table#FAT32
"The boot sector uses a 32-bit field for the sector count, limiting
the FAT32 volume size to 2 TB for a sector size of 512 bytes and 16 TB
for a sector size of 4,096 bytes."
* Microsoft: Default cluster size for NTFS, FAT, and exFAT / Default
cluster sizes for FAT32
https://support.microsoft.com/en-us/kb/140365
Trying to create a FAT32 file system in a partition larger than 2 TiB
results in unallocated space being left after the file system.
Nuances:
[1] Larger sector sizes allow larger maximum volume sizes up to 16 TiB
with 4096 byte sectors.
[2] mkdosfs/mkfs.fat has an -S SECTOR_SIZE option which allows changing
the "logical" sector size of the file system allowing the maximum
volume to be proportionally increased.
[3] mkfs.fat appears to have an signed overflow bug when the size of the
partition is larger than maximum signed 32-bit integer of logical(?)
sectors. (2 TiB for a sector size of 512 bytes). It reports the
partition size as minus size and creates a 1 TiB file system.
GParted wants a single maximum file system size and the code is not
ready for a differing maximum file system size for different sector
sizes.
In fat16::create() could specify larger "logical" sector sizes to
mkfs.fat when the partition is larger than 2 TiB to allow maximum volume
size to be increased further. However that will take a lot of cross
platform testing to ensure that all sorts of devices support "logical"
sector sizes other than 512 bytes on devices with a hardware sector size
of 512 bytes. This is too much effort.
Therefore implement a single FAT32 maximum volume size of 2 TiB.
Bug 763896 - GParted not restricting creation of FAT32 volume to maximum
size of 2 TiB
As part of the internal block copy operation 5 initial ranges of blocks
are copied using different block sizes to determine the fastest. Then
the remainder is copied using the fastest block size. Each of these
copies reports progress independently, so during the benchmarking phase
the progress bar flashes 5 times as it goes from 0 to 100% in a fraction
of a second, before showing the progress of the remainder.
This looks bad, so report a single progress bar for all the ranges of
blocks copied in a single copy operation.
Already have variables done and length which track progress within each
copied range; and total_done which records amount copied in previous
ranges. Just add total_length to allow overall progress to be reported.
Bug 762367 - Use a single progress bar for the whole of the internal
copy operation
Previously total_done was updated in copy_thread() after copying of the
blocks, but importantly before the last call to set_progress_info() to
update the progress bar. Having total_done varying during the copy of a
single range of blocks, single call to copy_blocks::copy(), is an
impediment to being able to report a single progress bar across the
whole internal copy operation.
Move updating of total_done to copy() immediately after copy_thread()
completes.
Bug 762367 - Use a single progress bar for the whole of the internal
copy operation
Copying of ntfs is performed using ntfsclone, which writes progress
indication to standard output like this:
# ntfsclone -f /dev/sdb2 /dev/sdb1 2> /dev/null
NTFS volume version: 3.1
Cluster size : 4096 bytes
Current volume size: 21474832384 bytes (21475 MB)
Current device size: 21474836480 bytes (21475 MB)
Scanning volume ...
100.00 percent completed
Accounting clusters ...
Space in use : 1832 MB (8.5%)
Cloning NTFS ...
100.00 percent completed
Syncing ...
Add ntfsclone progress tracker for ntfsclone command. Deliberately
doesn't stop the progress bar. See comment in ntfs::clone_progress()
for the explanation.
Bug 762366 - Add progress bar to NTFS file system specific copy method
Mostly the code is explicit and calls the emit() method when emitting a
signal [1], like this:
signal_name.emit();
However there are a few cases which use the function call operator on
the signal object [2], like this:
signal_name();
The behaviour is identical [3] but it is preferred to be explicit that a
signal callback is being initiated, and it also makes them much easier
to search for too.
[1] List explicit emit() signal calls
fgrep '.emit(' src/*.cc
[2] List function call operator emitted signals
egrep "`sed -n '/sigc::signal/s/.*sigc::signal.*> *\([a-zA-Z_]*\).*/\1/p' include/*.h | tr '\n' '|' | sed 's/\(.*\).$/[^a-zA-Z_](\1)\\\(/'`" src/*.cc
[3] Quote from the libsigc++ Reference Manual, class sigc::signal
https://developer.gnome.org/libsigc++/stable/classsigc_1_1signal7.html#ab37db0ecc788824d0baa3c301efc8dcd
result_type sigc::signal<...>::operator()(...)
Triggers the emission of the signal (see emit())
For non-progress tracked external commands the command being executed is
displayed in the Apply pending operations dialog, just below the top
pulsing progress bar. However for progress tracked external commands
the description of the parent operation detail is displayed instead.
Example 1: non-progress tracked xfs check repair:
TreePath
Check and repair file system (xfs) on /dev/sdc1 0
+ calibrate /dev/sdc1 0:0
+ check file system on /dev/sdc1 for errors and (if po... 0:1
+ xfs_repair -v /dev/sdc1 0:1:0
+ [empty stdout] 0:1:0:0
+ [stderr]Phase 1 - find and verify superblock... 0:1:0:1
"xfs_repair -v /dev/sdc1" (TreePath 0:1:0) is shown because it is the
latest updated operation detail which is timed (set to status
executing).
Example 2: progress tracked ext4 copy using e2image:
TreePath
Copy /dev/sdc2 to /dev/sdc3 0
+ calibrate /dev/sdc2 0:0
+ check file system on /dev/sdc2 for errors and (if po... 0:1
+ set partition type on /dev/sdc3 0:2
+ copy file system of /dev/sdc2 to /dev/sdc3 0:3
+ e2image -ra -p /dev/sdc2 /dev/sdc3 0:3:0
+ [stdout]Scanning inodes... 0:3:0:0
+ [stderr]e2image 1.42.9 (28-Dec-2013)... 0:3:0:1
"copy file system of /dev/sdc2 to /dev/sdc3" (TreePath 0:3) is shown
because that operation detail is also timed and it is being constantly
updated by the progress bar updates via it.
Change execute_command() to update the progress bar via the operation
detail it creates to hold the command being executed, instead of the
parent operation detail, to resolve the above. Also replaces calling
operationdetail.get_last_child() throughout the method.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Remove starting and stopping the progress bar in xfs::copy(). The
progress bar will be automatically started in xfs::copy_progress()
callback when run_progressbar() is called; and automatically stopped in
FileSystem::execute_command() when it calls stop_progress() at the end.
Note that this will now not initialise the progress bar from zero
immediately that the XFS copy is started, but instead 0.5 seconds into
the copy when xfs::copy_progress() timed callback is called for the
first time.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
The timed progress tracking callback for execution of xfs copy follows
this pattern:
sigc::connection c;
...
c = Glib::signal_timeout().connect( ... sigc::mem_fun( *this, &xfs::copy_progress ) ..., 500 /*ms*/ );
... execute_command( ... );
c.disconnect();
As with output progress tracking callbacks for ext2/3/4 and ntfs file
system specific commands, pass the callback slot and a flag into
execute_command() and connect the timed callback inside. This
simplified the pattern to:
... execute_command( ...|EXEC_PROGRESS_TIMED,
static_cast<TimedSlot>( sigc::mem_fun( *this, &xfs::copy_progress ) ) );
NOTE:
The type of sigc::mem_fun() doesn't allow the compiler to choose between
the two overloaded variants of execute_command() with the fourth
parameter of either (full types without typedefs of StreamSlot and
TimedSlot respectively):
sigc::slot<void, OperationDetail *> stream_progress_slot
sigc::slot<bool, OperationDetail *> timed_progress_slot
Therefore have to cast the result of all callback slots to the relevant
type. Hence:
static_cast<StreamSlot>( sigc::mem_fun( *this, &{CLASS}::{NAME}_progress ) )
static_cast<TimedSlot>( sigc::mem_fun( *this, &xfs::copy_progress ) )
References:
* [sigc] Functor not resolving between overloaded methods with
different slot types
https://mail.gnome.org/archives/libsigc-list/2016-February/msg00000.html
* Bug 306705 - Can't overload methods based on different slot<>
parameters.
https://bugzilla.gnome.org/show_bug.cgi?id=306705
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
For the relevant stream from a file system specific command being
tracked, there were 2 callbacks attached: update_command_output() and
update_command_progress(). When called, update_command_progress() just
emitted signal_progress to call the file system specific progress
tracker callback. Like this:
signal_update.emit() -> update_command_output()
-> update_command_progress()
signal_progress.emit() -> {CLASS}::{NAME}_progress()
Instead just connect the file system specific progress tracker callback
directly to signal_update and bypass the unnecessary
update_command_progress() method and the signal_progress signal. Like
this:
signal_update.emit() -> update_command_output()
-> {CLASS}::{NAME}_progress()
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
All the output progress tracking callbacks for execution of ext2/3/4 and
ntfs file system specific commands followed this pattern:
sigc::connection c = signal_progress.connect( sigc::mem_fun( *this, &ext2::..._progress ) );
bool success = ! execute_command( ... );
c.disconnect();
return success;
Instead, pass the callback slot and a flag into execute_command() and
connect the callback inside. This simplifies the pattern to:
return ! execute_command( ...|EXEC_PROGRESS_STDOUT,
sigc::mem_fun( *this, &ext2::..._progress ) );
Note that as the progress tracking callbacks are only registered against
updates to the relevant stream from the tracked commands they won't be
called when the other stream is updated any more.
Also note that signal_progress is a member of the FileSystem class and
derived objects so lives as long as GParted is running, therefore the
progress tracking callbacks need explicitly disconnecting at the end of
execute_command(). However signal_update is a member of the PipeCapture
class of which the output and error local variables in execute_command()
are types. Therefore there is no need to explicitly disconnect the
signal_update callbacks as they will be destructed along with the
callback slots when they go out of scope at the end of the
execute_command() method.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Remove unused members: fraction and progress_text from the
OperationDetail class now that the ProgressBar class has superseded
their use. This also allows removal of timer_global member from the
copy_blocks class. Timer_global was only used to track the elapsed time
copying blocks and allow the remaining time to be estimated and written
into progress_text. The ProgressBar class also does this itself
internally.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Most of the file system specific command progress trackers followed this
pattern:
void {CLASS}::{NAME}_progress( OperationDetail *operationdetail )
{
ProgressBar & progressbar = operationdetail->get_progressbar();
// parse output for progress and target values
if ( // have progress and target values )
{
if ( ! progressbar.running() )
progressbar.start( target );
progressbar.update( progress );
operationdetail->signal_update( *operationdetail );
}
else if ( // found progress finished )
{
if ( progressbar.running() )
progressbar.stop();
operationdetail->signal_update( *operationdetail );
}
}
That is a lot of repetition handling progress bar updates and
OperationDetail object update signalling. Remove the need for direct
access to the single ProgressBar object and provide these two
OperationDetail methods instead:
// Start and update in one
run_progressbar( progress, target, optional text_mode );
stop_progressbar();
Now the file system specific command progress trackers can become:
void {CLASS}::{NAME}_progress( OperationDetail *operationdetail )
{
// parse output for progress and target values
if ( // have progress and target values )
{
operationdetail->run_progressbar( progress, target );
}
else if ( // found progress finished )
{
operationdetail->stop_progressbar();
}
}
Make ProgressBar::get_progressbar() a private method to enforce use of
the new way to access the progress bar via the run_progress() and
stop_progressbar() methods. Then make the Dialog_Progress a friend
class to OperationDetail so that the Apply pending operations dialog can
still access the single ProgressBar object for its querying needs.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Using e2image to copy a file system looks like this. (Intermediate
progress lines which are constantly overwritten are indicated with ">").
# e2image -ra -p /dev/sdb4 /dev/sdb5
e2image 1.42.13 (17-May-2015)
Scanning inodes...
> Copying 0 / 276510 blocks (0%)
> Copying 8845 / 276510 blocks (3%)
> Copying 48433 / 276510 blocks (18%)
> Copying 77135 / 276510 blocks (28%)
> Copying 111311 / 276510 blocks (40%)
> Copying 137039 / 276510 blocks (50%)
> Copying 166189 / 276510 blocks (60%) 00:00:03 remaining at 108.20 MB/s
> Copying 190285 / 276510 blocks (69%) 00:00:03 remaining at 106.19 MB/s
> Copying 209675 / 276510 blocks (76%) 00:00:02 remaining at 102.38 MB/s
> Copying 238219 / 276510 blocks (86%) 00:00:01 remaining at 103.39 MB/s
> Copying 256692 / 276510 blocks (93%) 00:00:00 remaining at 100.27 MB/s
Copied 276510 / 276510 blocks (100%) in 00:00:10 at 108.01 MB/s
Note that the copying figures are reported in file system block size
units and the progress information is written to stderr, hence needing
these two previous commits:
Record file system block size where known (#760709)
Call any FS specific progress trackers for stderr updates too (#760709)
Add progress tracking function for e2image command. Also tracks when
the text progress indicator has passed in the output so that the
progress bar can be stopped as well as started when needed.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
So far the signal_progress callback slot was only emitted when standard
output from the file system specific command was updated. This was okay
as all the commands until now wrote their progress information to
stdout. However e2image writes its progress information to stderr,
therefore also emit signal_progress when stderr is updated too.
This does mean that the file system specific *_progress() tracking
callbacks will be called when either of the OperationDetail objects
containing stdout or stderr are updated. Therefore the trackers may be
called when there is no update to the stream from which it is parsing
the progress information. This is not a problem as the tracker will
just update the progress bar with the same information it already has.
Also it won't happen much as only e2image is known to write to both
streams, and then only one line to stdout and the updated progress
information to stderr. This is just an observation and not an issue
which needs coding around.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Record the file system block size in the Partition object. Only
implemented for file systems when set_used_sectors() method has already
parsed the value or can easily parse the value from the existing
executed command(s).
Needed for ext2/3/4 copies and moves performed using e2image so that
they can be tracked in bytes by the ProgressBar class as e2image reports
progress in file system block size units.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
XFS uses a file system specific method to copy the partition using
"xfsdump | xfsrestore". Monitor progress by periodically querying the
destination file system usage and comparing to the source file system
usage. Use 0.5 seconds as the polling interval to match that used by
the internal block copying algorithm.
NOTE:
The number of used blocks in the source and destination file system will
be very close but may not match exactly. I have seen an XFS copy finish
with the following progress text:
1.54 GiB of 1.50 GiB copied (-00:00:02 remaining)
Allow the progress bar to overrun like this as it is informing the user
that it actually copied a little more data and took a little longer than
expected. Needs these two previous commits to correctly round and
format the negative time remaining:
Fix rounding of negative numbers (#760709)
Fix formatting of negative time values (#760709)
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
... to display a negative sign before the hours, minutes and seconds.
Before:
Utils::format_time(-1) = "00:00:0-1"
Utils::format_time(-119) = "00:0-1:0-59"
After:
Utils::format_time(-1) = "-00:00:01"
Utils::format_time(-119) = "-00:01:59"
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Utils::round() was doing +0.5 then truncate. Correct for positive
values. Wrong for negative values.
E.G.
Utils::round(-1.4)
= trunc(-1.4 + 0.5)
= trunc(-0.9)
= 0
Round of -1.4 is definitely not 0. Fix this for negative values by
subtracting 0.5 then truncating.
Reference:
How can I convert a floating-point value to an integer in C?
https://www.cs.tut.fi/~jkorpela/round.html
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
When the ntfs resize operation had almost completed, percentage complete
was >= 99.9%, the progress tracker was passing 0.04 (4%) to the progress
bar. After reading the next chunk of output from the ntfsresize command
the last line contained this text:
" 4) set the bootable flag for the partit"
End of the ntfsresize command output for context:
Relocating needed data ...
100.00 percent completed
Updating $BadClust file ...
Updating $Bitmap file ...
Updating Boot record ...
Syncing device ...
Successfully resized NTFS on device '/dev/sdd4'.
You can go on to shrink the device for example with Linux fdisk.
IMPORTANT: When recreating the partition, make sure that you
1) create it at the same disk sector (use sector as the unit!)
2) create it with the same partition type (usually 7, HPFS/NTFS)
3) do not make it smaller than the new NTFS filesystem size
4) set the bootable flag for the partition if it existed before
Otherwise you won't be able to access NTFS or can't boot from the disk!
If you make a mistake and don't have a partition table backup then you
can recover the partition table by TestDisk or Parted's rescue mode.
This was occurring because *scanf() can't actually report failure to
match fixed text after conversion of the last variable. See code
comment in ntfs::resize_progress() for more details. Fix by using
.find() instead to match the required "percent completed" explicit text
of the progress information when it appears on the last line.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Adapt the ntfs resize progress tracker to use the new ProgressBar class.
Also make it track when the text progress indicator has passed in the
output so that the progress bar can be stopped as well as started when
needed.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Adapt the ext2 fsck progress tracker to use the new ProgressBar class.
Also make it track when the text progress bar has completely passed in
the output so that the progress bar can be stopped as well as started
when needed.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Adapt the ext2 create file system progress tracker to used the new
ProgressBar class. Also make it track when the text progress indicator
completes so that the progress bar can be stopped as well as started
when needed.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Adapt the ext2 resize progress tracker to the new ProgressBar class.
Also update the progress function to track when text progress bars have
completely passed in the output so that the progress bar can be stopped
as well as started when needed.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Change the Applying pending operations dialog so that it takes it source
of progress from the single ProgressBar object, rather than the fraction
value in every OperationDetail object. Also remove ProgressBar
debugging now that it is being used to drive the UI.
NOTE:
This temporarily causes the existing file system specific progress bars
to not be shown because they still update via the fraction member in
each OperationDetail object, rather than the new ProgressBar. This will
be corrected in following commits.
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
1) Multiple progress bars
The OperationDetail class contains member fraction which is used to feed
data to the current operation progress bar shown in the Applying pending
operations dialog. Dialog_Progress::on_signal_update() gets called for
every updated OperationDetail object and depending on whether fraction
is > 0.0 or not, switches between showing a growing or pulsing progress
bar. This leads to the conclusion that every OperationDetail object
currently being updated is effectively driving the single on screen
progress bar with different data.
The Copy_Blocks code is careful to update text and faction in a single
OperationDetail object and everything is good. The on screen progress
bar is switched into growing mode and then grows to 100%.
Since external command output is updated in real time [1] there are two
OperationDetail objects, one for stdout and one for stderr, which are
updated whenever data is read from the relevant stream. Also now that
progress is interpreted from some external command output [2][3][4] a
separate OperationDetail object is getting updated with the progress
fraction. (Actually the grandparent OperationDetail of the ones
receiving stdout and stderr updates as used by the file system specific
*_progress() methods). In the normal case of an external command
which is reporting it's progress two OperationDetails are constantly
being updated together, the OperationDetail object tracking stdout and
it's grandparent receiving progress fraction updates. This causes the
the code in Dialog_Progress::on_signal_update() to constantly switch
between growing and pulsing progress bar mode. The only reason this
doesn't flash the progress bar is because the stdout OperationDetail
object is updated first and before the 100 ms timeout fires to pulse the
bar, it's grandparent is updated with the new fraction to keep growing
the bar instead.
2) Common code
The Copy_Blocks code currently tracks the progress of blocks copied
against target amount, which it has to do anyway. That information is
then used to generate the text and fraction to update into the
OperationDetail object and drive the on screen progress bar. This same
level of tracking is wanted for the XFS and ext2/3/4 file system
specific copy methods.
Conclusion and solution
Having multiple sources of progress bar data is a problem and makes it
clear that there must be only one source of progress data. Also some
code can be shared for tracking the amount of blocks copied and
generating the display.
Therefore have a single ProgressBar object which is used everywhere.
This commit
It just creates a single ProgressBar object which is available via all
OperationDetail objects and Copy_Blocks is updated accordingly. Note
that the ProgressBar still contains debugging and that the GUI progress
bar of the current operation is still driven via the fraction member in
any OperationDetail object.
Referenced commits:
[1] 52a2a9b00a
Reduce threading (#685740)
[2] ae434579e1
Display progress for e2fsck (#467925)
[3] baea186138
Display progress for mke2fs (#467925)
[4] 57b028bb8e
Display progress during resize (#467925)
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Write a generic progress bar class. Has the following features:
* Has separate progress and target numbers, rather than a single
completion fraction, to enable the the next feature.
* Optionally generates text reporting the amount of data copied using
the progress and target numbers like this:
"1.00 MiB of 16.00 MiB copied"
* After running for 5 seconds, also add estimated remaining time.
(Waits to allow the data copying rate to settle down a little before
estimating the remaining time). Looks like this:
"1.00 MiB of 16.00 MiB copied (00:01:59) remaining)"
The ProgressBar class is not driving the visual progress bar yet. It
has just been added into the internal block copy algorithm and generates
debug messages showing the progress bar is operating correctly.
Debugging looks like this:
DEBUG: ProgressBar::start(target=2.0636e+09, text_mode=PROGRESSBAR_TEXT_COPY_BYTES)
DEBUG: ProgressBar::update(progress=1.30023e+08) m_fraction=0.0630081 m_text="124.00 MiB of 1.92 GiB copied"
DEBUG: ProgressBar::update(progress=2.67387e+08) m_fraction=0.129573 m_text="255.00 MiB of 1.92 GiB copied"
DEBUG: ProgressBar::update(progress=4.0475e+08) m_fraction=0.196138 m_text="386.00 MiB of 1.92 GiB copied"
...
DEBUG: ProgressBar::update(progress=1.13351e+09) m_fraction=0.549289 m_text="1.06 GiB of 1.92 GiB copied (00:00:04 remaining)"
DEBUG: ProgressBar::update(progress=1.26249e+09) m_fraction=0.611789 m_text="1.18 GiB of 1.92 GiB copied (00:00:04 remaining)"
DEBUG: ProgressBar::update(progress=1.39041e+09) m_fraction=0.67378 m_text="1.29 GiB of 1.92 GiB copied (00:00:03 remaining)"
...
DEBUG: ProgressBar::update(progress=1.97552e+09) m_fraction=0.957317 m_text="1.84 GiB of 1.92 GiB copied (00:00:00 remaining)"
DEBUG: ProgressBar::update(progress=2.0636e+09) m_fraction=1 m_text="1.92 GiB of 1.92 GiB copied"
DEBUG: ProgressBar::stop()
Bug 760709 - Add progress bars to XFS and EXT2/3/4 file system specific
copy methods
Simplify code in Dialog_Partition_Info::Dialog_Info() which was open
coding concatenating together a vector of strings with a new line
between each. Replace with Glib::build_path(), as used elsewhere in
this method and other code.
Create and open a LUKS mapping but don't create any file system within.
# cryptsetup luksFormat /dev/sdb5
# cryptsetup luksOpen /dev/sdb5 sdb5_crypt
GParted was incorrectly reporting this warning:
Unable to read the contents of this file system!
Because of this some operations may be unavailable.
The cause might be a missing software package.
The following list of software packages is required for luks file
system support: dmsetup.
This is even though the usage figures for the on disk LUKS encryption
are fully known. See luks::set_used_sectors().
This was occurring because derived PartitionLUKS::set_usage_known()
was checking usage figures for the outer LUKS and the inner encrypted
file system, unknown in this case. Correct when displaying figures in
the UI, but not correct in GParted_Core::set_used_sectors() when only
checking the outer LUKS usage figures were set correctly. Fix this.
Bug 760080 - Implement read-only LUKS support
At the moment any messages for an encrypted file system aren't shown,
only messages from the outer PartitionLUKS object are shown. Also in
Win_GParted::activate_paste() the selected Partition object, possibly
a derived PartitionLUKS, is cloned and the messages cleared.
Therefore a set of accessor methods must be provided to query and modify
partition messages. Messages will be stored in the Partition object to
which they are added and retrieved from all. So in the case of a
derived PartitionLUKS they will be retrieved from the messages vector of
the PartitionLUKS object itself and the messages vector for the
encrypted file system it contains.
To replace code like this in GParted_Core:
partition_temp->messages = messages;
We might naturally provide a set_messages() method which assigns the
messages vector and is used like this:
partition_temp->set_messages( messages );
However on a PartitionLUKS object what should set_messages() do? By the
name it will replace any existing messages in the PartitionLUKS object
itself, but what should happen to the messages for the contained
encrypted Partition object? Should they be cleared or left alone?
Rather than implement set_messages() with unclear semantics implement
append_messages(), which in the PartitionLUKS object case will clearly
leave any messages for the contained encrypted Partition object alone.
Append_messages() is then used to add messages as the Partition or
PartitionLUKS objects when populating the data in GParted_Core.
Bug 760080 - Implement read-only LUKS support
For LUKS formatted partitions add an encryption section into the
Information dialog and display the type of encryption, path, UUID and
status of the encryption.
The file system section continues to display appropriate file system
details, including the partition graphic with the file system specific
border colour and correct usage. The details will either be of a plain
file system, an encrypted file system, or nothing when there is no open
dm-crypt mapping, leaving the encrypted file system inaccessible.
Should there be LUKS encryption directly within LUKS encryption then the
details of the inner encryption will be displayed in the file system
section. However this configuration will not be further supported by
GParted.
Bug 760080 - Implement read-only LUKS support
There is already the set of methods in the Partition class to report the
file system usage. Virtualise them and provide PartitionLUKS specific
implementations to calculate the usage of a file system wrapped in LUKS
encryption.
See the ascii art and comment in PartitionLUKS.cc for the details of
those calculations.
Bug 760080 - Implement read-only LUKS support
LUKS headers don't provide any concept of label. Also there is already
the method Partition::get_filesystem_label() for getting the *file
system* label, so virtualise it and provide an appropriate
implementation to get the label of an encrypted file system represented
within a derived PartitionLUKS object. This causes the label to be
displayed correctly in the main window.
It also happens to display the encrypted file system label in the
Information dialog for a LUKS formatted partition. However the whole
Information dialog will be addressed differently in a following commit.
Bug 760080 - Implement read-only LUKS support
In the main window disk graphic, when there is an open dm-crypt mapping,
display the colour of the encrypted file system.
Bug 760080 - Implement read-only LUKS support
In the File System column in the GUI, when there is an open dm-crypt
mapping, display the colour square for the encrypted file system within
and the text as "[Encrypted] FSTYPE". For closed mappings nothing can
be known about the encrypted file system within so continue to display a
purple square and the text "[Encrypted]".
Looks like:
Partition | File System
...
/dev/sdb3 # ext4
v /dev/sdb4 * # extended
/dev/sdb5 # [Encrypted]
/dev/sdb6 * # [Encrypted] unknown
/dev/sdb7 * # [Encrypted] ext4
Bug 760080 - Implement read-only LUKS support
Partition object represents a region of a disk and the file system
within. GParted always displays the colour base of the type of the file
system. Therefore remove the color member and always look it up from
the type of the file system as needed.
This makes one less member that will need virtual accessor methods with
different handling in the derived PartitionLUKS class.
Bug 760080 - Implement read-only LUKS support
When there exists an open dm-crypt mapping, populate the encrypted
Partition object representing the encrypted file system.
Bug 760080 - Implement read-only LUKS support
Absolute minimum implementation of a PartitionLUKS class which can be
constructed, polymorphically copied and destroyed. Contains an
"encrypted" member of type Partition to represent the encrypted file
system within the LUKS format.
Create PartitionLUKS objects instead of base Partition objects when a
LUKS formatted partition is found. Only the base Partition object
member values have been populated, and the "encrypted" member remains
blank at this point.
Bug 760080 - Implement read-only LUKS support
This is the equivalent change as made to set_mountpoints() in an earlier
commit. Change GParted_Core::set_used_sectors() from being called with
a vector of partitions and processing them all to being called per
partition. This is in preparation for calling set_used_sectors() on a
single Partition object inside a PartitionLUKS object.
Bug 760080 - Implement read-only LUKS support
Populate the canonical device name, /dev/mapper/NAME, used to access the
encrypted file system into the mount points of the Partition object.
This is the equivalent of what is already done for the Volume Group name
and SWRaid Array device.
This does get displayed in the Mount Point column in the main window,
which isn't wanted. However the data will be needed when displaying
details of the encryption mapping in the Information dialog. Both will
be dealt with in following commits.
Bug 760080 - Implement read-only LUKS support
Previously GParted_Core::set_mountpoints() was called with a vector of
partitions and processed them all. Now make set_mountpoints() process a
single partition and push the calls to it down one level from
set_devices_thread() into set_device_partitions() and
set_device_one_partition(). This is in preparation for having an
encrypted file system represented as a Partition object inside a
PartitionLUKS object and needing to call set_mountpoints() for the inner
single Partition object.
Bug 760080 - Implement read-only LUKS support
Only load the LUKS_Info cache of active dm-crypt mappings when the first
LUKS partition is encountered. Not needed from a performance point of
view as the longest that I have ever seen "dmsetup table --target crypt"
take to run is 0.05 seconds. Just means that the dmsetup command is
only run when there are LUKS partitions and the information is needed.
Bug 760080 - Implement read-only LUKS support
Populate the used, unused and unallocated figures in the Partition
object for a LUKS formatted partition. See comment in
luks::set_used_sectors() for the rational of what is used, unused and
unallocated.
As that rational mentions, a LUKS header does not store the size of the
encrypted data and is assumed to extend to the end of the partition by
the tools which start the mapping.
An underlying block device of 128 MiB (131072 KiB).
# sfdisk -s /dev/sde
131072
An active LUKS mapping at offset 2 MiB (4096 512-byte sectors) and
length 126 MiB (129024 KiB, 258048 512-byte sectors).
# sfdisk -s /dev/mapper/sde_crypt
129024
# cryptsetup status sde_crypt
/dev/mapper/sde_crypt is active.
type: LUKS1
cipher: aes-cbc-essiv:sha256
keysize: 256 bits
device: /dev/sde
offset: 4096 sectors
size: 258048 sectors
mode: read/write
No size/length reported when dumping the LUKS header, just (payload)
offset.
# cryptsetup luksDump /dev/sde
LUKS header information for /dev/sde
Version: 1
Cipher name: aes
Cipher mode: cbc-essiv:sha256
Hash spec: sha1
Payload offset: 4096
MK bits: 256
MK digest: 7f fb ba 40 7e ba e4 3b 2f c6 d0 93 7b f7 05 49 7b 72 d4 ad
MK salt: 4a 5b 54 f9 7b 67 af 6e ef 16 31 0a fe d9 7e 5f
c3 66 dc 8a ed e0 07 f4 45 c3 7c 1a 8d 7d ac f4
MK iterations: 37750
UUID: 0a337705-434a-4994-a842-5b4351cb3778
...
Shrink the LUKS mapping to 64 MiB (65536 KiB, 131072 512-byte sectors).
# cryptsetup resize --size 131072 sde_crypt
# sfdisk -s /dev/mapper/sde_crypt
65536
# cryptsetup status sde_crypt
/dev/mapper/sde_crypt is active.
type: LUKS1
cipher: aes-cbc-essiv:sha256
keysize: 256 bits
device: /dev/sde
offset: 4096 sectors
size: 131072 sectors
mode: read/write
Stop and start the LUKS mapping.
# cryptsetup luksClose sde_crypt
# cryptsetup luksOpen /dev/sde sde_crypt
The size of the LUKS mapping is back to 126 MiB (129024 KiB, 258048
512-byte sectors), extending to the end of the partition.
# sfdisk -s /dev/mapper/sde_crypt
129024
# cryptsetup status sde_crypt
/dev/mapper/sde_crypt is active.
type: LUKS1
cipher: aes-cbc-essiv:sha256
keysize: 256 bits
device: /dev/sde
offset: 4096 sectors
size: 258048 sectors
mode: read/write
Bug 760080 - Implement read-only LUKS support
Also load the starting offset and length of the active dm-crypt mapping
into the LUKS_Info module from the dmsetup output. This provides the
location and size of the encrypted data within the underlying block
device.
Note that dmsetup reports in units of 512 bytes sectors [1], the GParted
LUKS_Info module uses bytes and GParted Partition objects work in device
sector size units. However the actual sector size of a dm-crypt mapping
[2] is the same as that of the underlying block device [3].
# modprobe scsi_debug dev_size_mb=128 sector_size=4096
# fgrep scsi_debug /sys/block/*/device/model
/sys/block/sdd/device/model:scsi_debug
# parted /dev/sde print
Error: /dev/sde: unrecognised disk label
Model: Linux scsi_debug (scsi)
Disk /dev/sde: 134MB
[3] Sector size (logical/physical): 4096B/4096B
Partition Table: unknown
# cryptsetup luksFormat /dev/sde
# cryptsetup luksOpen /dev/sde sde_crypt
# parted /dev/mapper/sde_crypt print
Error: /dev/mapper/sde_crypt: unrecognised disk label
Model: Linux device-mapper (crypt) (dm)
Disk /dev/mapper/sde_crypt: 132MB
[2] Sector size (logical/physical): 4096B/4096B
Partition Table: unknown
# cryptsetup status sde_crypt
/dev/mapper/sde_crypt is active.
type: LUKS1
cipher: aes-cbc-essiv:sha256
keysize: 256 bits
device: /dev/sde
offset: 4096 sectors
size: 258048 sectors
mode: read/write
# dmsetup table --target crypt
...
sde_crypt: 0 258048 crypt aes-cbc-essiv:sha256 0000000000000000000000000000000000000000000000000000000000000000 0 8:64 4096
[1] Both cryptsetup and dmsetup report the offset as 4096 and the size/
length as 258048. 128 MiB / (4096+258048) = 512 byte units, even on a
4096 byte sector size device.
Update debugging of LUKS to this:
# ./gpartedbin
======================
libparted : 2.4
======================
DEBUG: /dev/sdb5: LUKS closed
DEBUG: /dev/sdb6: LUKS open mapping /dev/mapper/sdb6_crypt, offset=2097152, length=534773760
/dev/sde: unrecognised disk label
DEBUG: /dev/sde: LUKS open mapping /dev/mapper/sde_crypt, offset=2097152, length=132120576
Bug 760080 - Implement read-only LUKS support
The code currently allows attempting to mount and unmount a LUKS
partition. It is nonsense to directly try to mount and unmount a LUKS
partition and obviously doesn't work. For read-only LUKS support there
is no need to attempt to apply this to the encrypted file system within.
Therefore prevent these operations for LUKS partitions.
Bug 760080 - Implement read-only LUKS support
This patchset is adding read-only LUKS support. Creation of LUKS is
planned to be a tick box adding encryption in the Create New Partition
dialog. Therefore remove the greyed out crypt-luks entry in the Create
New Partition dialog and the Format menu.
Bug 760080 - Implement read-only LUKS support
Provide a minimal implementation of a luks file system class which only
does busy detection.
NOTE:
For now, read-only LUKS support, a LUKS partition will be busy when a
dm-crypt mapping exists. Later when read-write LUKS support is added
GParted will need to look at the busy status of the encrypted file
system within the open LUKS partition and map LUKS partition busy status
to encryption being open or closed.
Bug 760080 - Implement read-only LUKS support
Load basic details of active Device-mapper encryption mappings from the
kernel. Use dmsetup active targets.
# cryptsetup luksFormat /dev/sdb5
# cryptsetup luksFormat /dev/sdb6
# cryptsetup luksOpen /dev/sdb6 sdb6_crypt
# ls -l /dev/mapper/sdb6_crypt /dev/dm-0
lrwxrwxrwx. 1 root root 7 Nov 15 09:03 /dev/mapper/sdb6_crypt -> ../dm-0
brw-rw----. 1 root disk 253, 0 Nov 15 09:03 /dev/dm-0
# ls -l /dev/sdb6
brw-rw----. 1 root disk 8, 22 Nov 15 09:02 /dev/sdb6
# dmsetup table --target crypt
sdb6_crypt: 0 1044480 crypt aes-cbc-essiv:sha256 0000000000000000000000000000000000000000000000000000000000000000 0 8:22 4096
So far just load the mapping name and underlying block device reference
(path or major, minor pair).
Note that all supported kernels appear to report the underlying block
device as major, minor pair in the dmsetup output. Underlying block
device paths are added to the cache when found during a search to avoid
stat(2) call on subsequent searches for the same path.
Prints debugging to show results, like this:
# ./gpartedbin
======================
libparted : 2.4
======================
DEBUG: /dev/sdb5: LUKS closed
DEBUG: /dev/sdb6: LUKS open mapping /dev/mapper/sdb6_crypt
Bug 760080 - Implement read-only LUKS support
Renamed from DEV_MAP_PATH to DEV_MAPPER_PATH. Moved so that the
constant is logically intended for use outside of the DMRaid class.
Also specifically make the string constant have external linkage, rather
than the default internal (static) linkage for constants, so that there
is only one copy of the variable in the program, rather than one copy in
each compilation unit which included DMRaid.h. Namely DMRaid.cc and
GParted_Core.cc.
References:
[1] Proper way to do const std::string in a header file?
http://stackoverflow.com/questions/10201880/proper-way-to-do-const-stdstring-in-a-header-file
[2] What is external linkage and internal linkage in C++
http://stackoverflow.com/questions/1358400/what-is-external-linkage-and-internal-linkage-in-c/1358796#1358796
Bug 760080 - Implement read-only LUKS support
History:
1) The constructor was added by commit:
6d8b169e73
2006-03-14 21:37:47
changed the way devices and partitions store their devicepaths. Instead of
2) Removed from most of the file system specific ::Copy() methods by
commit:
ad9f2126e7
2006-03-19 15:30:20
fixed issues with copying (see also #335004) cleanups + added FIXME added
3) Removed from GParted_Core::copy() method by commit:
7bb7e8a84f
2006-05-23 22:17:34
Use ped_device_read and ped_device_write instead of 'dd' to copy
4) Finally removed from the last place in xfs::Copy() method by commit:
e414b71b73
2012-01-11 19:49:13
Update xfs resize and copy to use new helper functions
The Partition(path) constructor is no longer used. Remove.
The sector_size parameter is unnecessary as the value can be retrieved
from the sector size of the selected Partition object on which the
create new, copy & paste or resize/move operation is being performed.
For the create new and resize/move operations it is trivial as the
existing unallocated or in use Partition object on which the operation
is being perform already contains the correct sector size. For the copy
& paste operation, which can copy across disk devices of different
sector sizes, we merely have to use the sector size of the existing
selected (destination) Partition object rather than copied (source)
Partition object. Hence these relevant lines in the new code:
Dialog_Partition_Copy::set_data(selected_partition, copied_partition)
new_partition = copied_partition.clone();
...
new_partition->sector_size = selected_partition.sector_size;
The expressions used in the call to Set() were comparing
lp_partition->type to 0 for the type parameter and passing it as a bool
for the inside_extended parameter. Libparted lp_partition->type is
actually an enumeration. The code was only working because of the
specific values assigned to the symbolic names, PED_PARTITION_NORMAL = 0
and PED_PARTITION_EXTENDED is non-zero (true).
Make the code use the symbolic names and not depend on the actual
enumeration values, which should be considered changeable and private to
libparted.
When displaying an unallocated partition
Win_GParted::set_valid_operations() calls GParted_Core::get_fs() with
parameter FS_UNALLOCATED.
Before this change, get_fs() would fail to find file system capabilities
set for FS_UNALLOCATED and construct a not supported capabilities set
and return that.
Afterwards, find_supported_filesystems() creates a not supported
capabilities set from the NULL pointer for FS_UNALLOCATED and adds this
entry into the FILESYSTEMS vector. Then get_fs() finds that not
supported capabilities set for FS_UNALLOCATED in the FILESYSTEMS vector
and returns that.
This makes no functional difference. It just seems right as other
unsupported but used file system types have entries in FILESYSTEM_MAP.
The struct FS constructor initialised every member *except* filesystem
and busy. Then in *most* cases after declaring struct FS, assignments
followed like this:
FS fs;
fs.filesystem = FS_BTRFS;
fs.busy = FS::GPARTED;
But member busy wasn't always initialised.
Add initialisation of members filesystem and busy to the struct FS
constructor. Specify optional parameter to the constructor to set the
filesystem member, or when left off filesystem is initialised to
FS_UNKNOWN.
get_fs() used to work by (1) returning the supported capabilities of the
requested file system found in the FILESYSTEMS vector; (2) if not found
return the supported capabilities for file system FS_UNKNOWN; and (3)
if that wasn't found either, create a not supported capabilities set for
FS_UNKNOWN and return that.
This is more complicated that required. Also the not supported
capabilities set, as created by struct FS() constructor, is the same as
that created in file_supported_filesystems() local variable fs_notsupp.
Simplify get_fs() just using a single not found code path returning a
not supported capabilities set.
Final step for full polymorphic handling of Partition objects is to
implement a virtual copy constructor. C++ doesn't directly support
virtual copy constructors, so instead use the virtual copy constructor
idiom [1]. (Just a virtual method called clone() which is implemented
in every polymorphic class and creates a clone of the current object and
returns a pointer to it).
Then replace all calls to the (monomorphic) Partition object copy
constructor throughout the code, except in the clone() implementation
itself, with calls to the new virtual clone() method "virtual copy
constructor".
Also have to make the Partition destructor virtual too [2][3] so that
the derived class destructor is called when deleting using a base class
pointer. C++ supports this directly.
[1] Wikibooks: More C++ Idioms / Virtual Constructor
https://en.wikibooks.org/wiki/More_C%2B%2B_Idioms/Virtual_Constructor
[2] When to use virtual destructors?
http://stackoverflow.com/questions/461203/when-to-use-virtual-destructors
[3] Virtuality
Guideline #4: A base class destructor should be either public and
virtual, or protected and nonvirtual
http://www.gotw.ca/publications/mill18.htm
Bug 759726 - Implement Partition object polymorphism
SQUASH: When first using pointers to Partition and calling delete
Copy assignment of Partition objects is now only performed in a few
places in the Operation and OperationResizeMove classes when updating
the displayed PartitionVector. (From Refresh_Visual() when each
operation is visually applied to the display_partitions vector; the
new_partition from the operation is copy assigned over the top of the
relevant existing partition in the display_partitions vector).
In general polymorphic copy assignment is complicated [1], and is now
unnecessary given the above limited use. All that is needed is a way to
polymorphically replace one Partition object with another in a
PartitionVector.
First, prevent further use of Partition object copy assignment by
providing a private declaration and no implementation, so the compiler
enforces this. Second implement and use PartitionVector method
replace_at() which replaces a pointer to one Partition object with
another at the specified index in the PartitionVector.
[1] The Assignment Operator Revisited
[Section:] Virtual assignment
http://icu-project.org/docs/papers/cpp_report/the_assignment_operator_revisited.html
Bug 759726 - Implement Partition object polymorphism
Now use a pointer to the Partition object in Dialog_Base_Partition class
and derived classes, Dialog_Partition_{Copy,New,Resize_Move}. This is
equivalent to how the Partition objects are managed in the Operation and
derived classes.
The Partition object is allocated and copy constructed in each derived
classes' set_data() method, called from each constructor and deallocated
in the destructors. Considering the remaining Big 3, these classes are
never copy constructed or copy assigned so provide private definitions
and no implementations so the compiler enforces this.
Bug 759726 - Implement Partition object polymorphism
The code in on_view_clicked() copy constructed a Partition object and
then in the following 3 lines only read a couple of public member
variables from the new copy.
Making a copy of the partition is unnecessary. Change to just creating
a constant reference to the Partition instead.
(It would also be an impediment to polymorphically using Partition
objects, except for the fact that gpart doesn't recognise LUKS
signatures so will never have to create a PartitionLUKS object).
Bug 759726 - Implement Partition object polymorphism
A number of methods in GParted_Core and Win_GParted were using local
Partition objects. Change them into pointers so that Partition object
polymorphism can be implemented.
Bug 759726 - Implement Partition object polymorphism
Change Win_GParted::copied_partition from Partition object which is
copied by value into a pointer to a Partition object object which is
allocated, copy constructed and deleted. Required as part of the
polymorphic implementation of Partitions.
As before when managing the lifetime of pointers to objects in a class
the Big 3 of destructor, copy constructor and copy assignment operator
need to be considered. A destructor is added to finally delete
copied_partition. A single Win_GParted object is only ever created and
destroyed in main(). The class is never copy constructed or copy
assigned. Make the compiler enforce this with private declarations and
no implementations.
Bug 759726 - Implement Partition object polymorphism
Operation classes now internally use pointers to Partition objects and
take on management of their lifetimes. As before, with the
PartitionVector class, when storing pointers in a class the Big 3 of
destructor, copy constructor and copy assignment operator also have to
be considered.
First, all the Partition objects are allocated in the derived Operation*
class parameterised constructors and freed in the associated
destructors. However the Operation classes are never copy constructed
or copy assigned; they are only ever created and destroyed. Only
pointers to the derived Operations are copied into the vector of pending
operations. Therefore the copy construtor and copy assignment operator
aren't needed. To enforce this provide inaccessible private
declarations without any implementation so that the compiler will
enforce this [1][2].
This example code fragment:
1 OperationCheck o1( device, partition );
2 OperationCheck o2 = o1;
3 o2 = o1;
Does these OperationCheck calls:
1 Implemented parameterised construtor,
2 Disallowed copy constructor,
3 Disallowed copy assignment
Trying to compile the above code would fail with errors like these:
../include/OperationCheck.h: In member function 'void GParted::Win_GParted::activate_check()':
../include/OperationCheck.h:36:2: error: 'GParted::OperationCheck::OperationCheck(const GParted::OperationCheck&)' is private
OperationCheck( const OperationCheck & src ); // Not implemented copy constructor
^
test.cc:2:21: error: within this context
OperationCheck o2 = o1;
^
../include/OperationCheck.h:37:19: error: 'GParted::OperationCheck& GParted::OperationCheck::operator=(const GParted::OperationCheck&)' is private
OperationCheck & operator=( const OperationCheck & rhs ); // Not implemented copy assignment operator
^
test.cc:3:4: error: within this context
o2 = o1;
^
[1] Disable copy constructor
http://stackoverflow.com/questions/6077143/disable-copy-constructor
[2] Disable compiler-generated copy-assignment operator [duplicate]
http://stackoverflow.com/questions/7823845/disable-compiler-generated-copy-assignment-operator
Bug 759726 - Implement Partition object polymorphism
The Operation classes contain partition objects which are copied by
value. Need to replace these with pointers to Partition objects instead
and manage their lifetimes so that they can be used polymorphically.
First step is to protect the partition members partition_new,
partition_original, and for OperationCopy class only, partition_copied
within the Operation classes and provide accessor methods.
get_partition_new() and get_partition_original() accessors are
implemented in the Operation base class so all derived classes get an
implementation. get_partition_new() is also virtual so that
OperationCheck and OperationDelete can override the implementation and
assert that they don't use partition_new. get_partition_copied() is
provided for the OperationCopy class only so can only be accessed via an
OperationCopy type variable.
Bug 759726 - Implement Partition object polymorphism
Remove PartitionVector push_back() and insert() methods which copy
construct Partitions objects into the vector. All the code has already
been changed to dynamically allocate Partition objects and use the
adoption variants of these methods named, push_back_adopt() and
insert_adopt(). Remove the no longer used methods.
Bug 759726 - Implement Partition object polymorphism
Replace all the current code which uses push_back() and insert() of a
local Partition object and gets it copy constructed into a
PartitionVector. Instead allocate a Partition object on the heap and
adopt a pointer into the PartitionVector using push_back_adopt() and
insert_adopt().
Bug 759726 - Implement Partition object polymorphism
GParted_Core and Operation classes both have an insert_unallocated()
method which do the same thing with very nearly identical code. Both
methods insert unallocated partitions into the vector of partitions
within the specified range of sectors to fill in any gaps larger than
1 MiB. The only difference was how the two methods got the device path;
the GParted_Core class method got it via a parameter and the Operation
class method got it by calling get_path() on its device member variable.
The GParted_Core insert_unallocated() method gets called during device
scanning and the Operation one gets called when constructing the visual
for a pending operation.
Consolidate down to a single insert_unallocated() implementation by
making the Operation class method call the GParted_Core class method.
Make the GParted_Core class method static and public so that it can be
called using the class name from outside the class.
Bug 759726 - Implement Partition object polymorphism
The current code uses push_back() and insert() to copy Partition objects
into the vector of pointers. This has a few issues:
1) Unnecessary copying of Partition objects;
2) Hides the nature of the PartitionVector class as a manager of
pointers to Partition objects by providing copy semantics to add
items. It is generally better to be explicit;
3) C++ doesn't provide polymorphic copy construction directly, but this
is easily worked around by following the Virtual Constructor idiom
[1], which would allow PartitionLUKS derived class objects to be
copied into the vector.
Add push_back_adopt() and insert_adopt() methods which add a pointer to
a Partition object into the PartitionVector adopting ownership.
[1] Wikibooks: More C++ Idioms / Virtual Constructor
https://en.wikibooks.org/wiki/More_C%2B%2B_Idioms/Virtual_Constructor
Bug 759726 - Implement Partition object polymorphism
The PartitionVector class is now internally using pointers to Partition
objects and taking on management of their lifetimes. It therefore has
to implement the Big 3: destructor, copy constructor and copy assignment
operator [1][2]. This is because the implicitly-defined copy
constructor and assignment operator perform memberwise "shallow copying"
and the destructor does nothing. This not correct for classes which
contain non-class types such as raw pointers.
The semantics of the interface still copies each Partition object into
the PartitionVector when they are added with push_back() and insert().
Note that a PartitionVector object is explicitly copy assigned in
Win_GParted::Refresh_Visual(). They are also implicitly copied when
(1) the implementing vector is resized larger to allow it to hold more
pointers to Partition objects than it previously had capacity for; and
(2) a Partition object is copied including the logicals PartitionVector
member.
[1] The rule of three/five/zero
http://en.cppreference.com/w/cpp/language/rule_of_three
[2] Rule of Three
https://en.wikipedia.org/wiki/Rule_of_three_%28C%2B%2B_programming%29
Bug 759726 - Implement Partition object polymorphism
Lots of files which use the Partition class relied on the declaration
being included via other header files. This is bad practice.
Add #include "Partition.h" into every file which uses the Partition
class which doesn't already include it. Header file #include guards are
specifically to allow this.
get_partitions() method was returning a vector of partitions. However
the calling code only needed to know whether any partitions were found
or not. Replace with found_partitions() method reporting the needed
boolean.
Now use of std::vector<Partition> partitions is hidden within the
Dialog_Rescue_Data class implementation.
Bug 759726 - Implement Partition object polymorphism
GParted was also displaying the GPT partition names as the file system
labels for some type of file systems.
Create 2 empty partitions on a GPT partitioned disk and name them like
this. Then create a hfsplus file system with an empty label. (Actually
single space character but blkid treats it as empty. Can't use GParted
for this because when no label is specified mkfs.hfsplus sets the label
to "untitled").
# sgdisk -p /dev/sdb
/dev/sdb: 4194304 sectors, 2.0 GiB
Logical sector size: 512 bytes
...
Number Start (sector) End (sector) Size Code Name
1 2048 526335 256.0 MiB 8300 empty partition
2 526336 1050623 256.0 MiB 8300 hfsplus partition
# mkfs.hfsplus -v " " /dev/sdb2
Even though only the GPT partition names are set to "SOMETHING
partition", GParted will display them as the file system labels too.
Also in the Information dialog for the empty partition a file system
UUID is displayed as well even though none exists.
Blkid output looks like this:
# blkid -V
blkid from util-linux 2.27.1 (libblkid 2.27.0, 02-Nov-2015)
# blkid | grep /dev/sdb
/dev/sdb1: PARTLABEL="empty partition" PARTUUID="bf3d2085-65b7-4ae6-97da-5ff968ad7d2c"
/dev/sdb2: UUID="2c893037-ff76-38f2-9158-2352ef5dc8de" TYPE="hfsplus" PARTLABEL="hfsplus partition" PARTUUID="457e9c2b-e4f2-4235-833b-28208592aaac"
With blkid from util-linux >= 2.22, released September 2012, it is
including additional PARTLABEL and PARTUUID name and value pairs for
GPT partitions [1]. The FS_Info module regular expressions used to
match the file system LABEL and UUID names also happen to match these
new PARTLABEL and PARTUUID names too. Hence this bug when GParted has
to fall back to using the FS_Info module to read the file system label,
when there is no working file system specific method. Effects: exfat,
f2fs, hfs, hfsplus, ufs, unknown.
Fix by requiring all the regular expressions used to search the blkid
output to also match the space character in front of the name.
[1] Util-linux 2.22 Release Notes
https://git.kernel.org/cgit/utils/util-linux/util-linux.git/tree/Documentation/releases/v2.22-ReleaseNotes?h=v2.22
Bug 759972 - GParted displays partition names also as file system labels
with new blkid for some file systems
In the UI new partitions were being named "unallocated" instead of
"New Partition #1", etc. Also deleting a new partition not yet created
deletes all new partitions from the operation list because it matches
by name only and they were all named "unallocated". Broken by this
recent commit:
762cd1094a
Return class member from Dialog_Partition_New::Get_New_Partition() (#757671)
Prior to this commit in the create new partition dialog code did these
relevant steps:
Dialog_Partition_New::Get_New_Partition()
Partition part_temp;
...
part_temp.Set(..., "New Partition #%1", ...);
Create local Partition object using default constructor which calls
Partition::Reset() and clears the paths vector. It then calls Set()
which appends the new name making the vector:
paths = ["New Partition #%1"]
After the above commit the code did these steps:
Dialog_Partition_New::Dialog_Partition_New(..., selected_partition, ...)
set_data(..., selected_partition, ...);
new_partition = selected_partition;
Dialog_Partition_New::Get_New_Partition(...)
new_partition.Set(..., "New Partition #%1", ...);
New_partition is copied from the selected unallocated partition in which
the new partition will be created. So new_partition is now named
"unallocated". Then the Set() call appends the new name making the
vector:
paths = ["unallocated", "New Partition #%1"]
As get_path() returns the first name in the paths vector the path name
changed from "New Partition #%1" to "unallocated" causing this bug.
Fix by resetting the new_partition object to clear all vestiges of it
being a copy of the selected unallocated partition, Reset() call, before
then calling Set(). This then appends the new name to an empty vector
making it contain just the required new name:
paths = ["New Partition #%1"]
Bug 759488 - Pending create partitions are all getting named
"unallocated"
Capture and parse the progress reports of ntfsresize and resize2fs and
update the dialog progress bar.
Bug 467925 - gparted: add progress bar during operation
Return newly constructed partition object by reference rather than by
copy from the Copy, Resize/Move and New dialog classes. This is another
case of stopping copying partition objects in preparation for using
polymorphic Partition objects. In C++ polymorphism has to use pass by
pointer and reference and not pass by value, copying, to avoid object
slicing.
The returned reference to the partition is only valid until the dialog
object containing the new_partition member is destroyed. This is okay
because in all three cases the returned referenced partition is copied
into a context with new lifetime expectations before the dialog object
is destroyed.
Case 1: GParted_Core::activate_paste()
Referenced new_partition is copied in the OperationCopy constructor
before the dialog object goes out of scope.
Operation * operation = new OperationCopy( ...,
dialog.Get_New_Partition( ... ),
... );
Case 2: GParted_Core::activate_new()
Referenced new_partition is copied in the OperationCreate
constructor before the dialog object goes out of scope.
Operation * operation = new OperationCreate( ...,
dialog.Get_New_Partition( ... ) );
Case 3: GParted_Core::activate_resize()
Temporary partition object is copied from the referenced
new_partition before the dialog object goes out of scope.
Partition part_temp = dialog.Get_New_Partition( ... );
Bug 757671 - Rework Dialog_Partition_New::Get_New_Partition() a bit
Replace the use of local Partition variable with class member in
preparation for Dialog_Partition_New::Get_New_Partition() being changed
to return the new partition object by reference instead of by value.
part_temp -> new_partition
Bug 757671 - Rework Dialog_Partition_New::Get_New_Partition() a bit
When creating a new extended partition, the unallocated space within it
was being created before adjusting the partition boundaries for
alignment reasons. This must be wrong. Move creation of the
unallocated space to after adjusting the partition boundaries for
alignment reasons. First introduced by this commit:
a30f991ca5
Fix size reduced by 1 MiB when created after cylinder aligned partition
Also added a big FIXME comment explaining how further adjustments are
still made by snap_to_alignment() to the partition boundaries including
a test case where this too late adjustment causes overlapping boundaries
and apply to fail.
Bug 757671 - Rework Dialog_Partition_New::Get_New_Partition() a bit
This is just to make the parameter name in the Dialog_Partition_New
constructor and set_data() method match the name of the equivalent
parameter in the Dialog_Partition_Copy and Dialog_Partition_Resize_Move
classes. (All three classes inherit from Dialog_Base_Partition and have
similar interfaces).
Bug 757671 - Rework Dialog_Partition_New::Get_New_Partition() a bit
Blkid recognises many more file system types and RAID member signatures
than libparted. GParted already uses blkid detection instead of or
before libparted for whole disk devices [1] and for ext4 detection [2]
(only required with libparted < 1.9.0). Also GParted could only use
blkid detection on non-512 byte sector devices [3] before libparted was
fixed in version 3.2 [4]. Blkid was documented as a mandatory
requirement from GParted 0.24.0 [5].
Util-linux package, of which blkid command is a part, is a core piece of
Linux software which is very actively maintained and used by lots of
other packages. Parted package is much less active and has added
detection of fewer file systems and doesn't recognise any RAID members.
In cases of multiple signatures within a partition blkid and libparted
can report different results leading to confusion and issues for
GParted. This was the primary reason for bug 688882 "Improve clearing
of file system signatures" and a number of other changes to GParted.
Also as the mount command links with libblkid it uses it's detection
when telling the kernel the type of a file system to be mounted.
There aren't any current issues with GParted's file system detection but
given the above argument, switch to using blkid before libparted for
file system detection. Only falling back to libparted when blkid
doesn't report a result, notably for extended partitions. Order of
information sources for detection is now:
1) mdadm (for SWRaid members)
2) blkid
3) libparted
4) GParted internal code
References:
[1] f8faee6377
Avoid whole disk FAT being detected as MSDOS partition table (#743181)
[2] 533eb1bc03
Added support for ext4 file systems
[3] 9e5e9f5627
Enhance file system detection to use FS_Info method - blkid
[4] http://git.savannah.gnu.org/cgit/parted.git/commit/?id=80678bdd957cf49a9ccfc8b88ba3fb8b4c63fc12
Fix filesystem detection on non 512 byte sectors
[5] 749a249571
Document blkid command as a mandatory requirement (#753436)
Bug 757781 - Always use blkid file system detection before libparted
The GParted_Core code only interacts with derived Operation* objects
through the Operation base class interface and in one case the
OperationCopy class. Therefore include Operation.h and OperationCopy.h
headers and no others.
ZFS labels were not cleared by GParted when clearing old file system
signatures.
# wget https://git.kernel.org/cgit/utils/util-linux/util-linux.git/plain/tests/ts/blkid/images-fs/zfs.img.bz2
# bzip2 -dc zfs.img.bz2 > /dev/sdb1
[In GParted format to cleared /dev/sdb1]
# blkid /dev/sdb1
/dev/sdb1: LABEL="tank" UUID="1782036546311300980" UUID_SUB="13179280127379850514" TYPE="zfs_member"
Update to also zero all 4 ZFS labels.
NOTE:
GParted now writes a little over 1 MiB when clearing old file system
signatures. As this is performed in the main thread the UI is not able
to respond during this action. Testing this on a range of USB flash
keys and hard drives found the slowest normal time to write this was
0.25 seconds, with an occasional outlier up to 2.8 seconds from a USB
flash key. This is considered acceptable.
Following the previous commit "Add erasing of SWRaid metadata 0.90 and
1.0 super blocks (#756829)" signature zeroing specified to write 4 KiB
of zeros at position end - 64 KiB, aligned to 64 KiB. Example operation
details from formatting a 1 GiB partition to cleared:
Format /dev/sdb8 as cleared
+ calibrate /dev/sdb8
+ clear old file system signatures in /dev/sdb8
+ write 68.00 KiB of zeros as byte offset 0
+ wite 4.00 KiB of zeros at byte offset 67108864
+ wite 64.00 KiB of zeros at byte offset 1073676288
+ write 8.00 KiB of zeros at byte offset 1073733632
+ flush operating system cache of /dev/sdb
However it actually wrote 64 KiB. This is because the rounding /
alignment was also applied to the zeroing length. Before this commit
rounding / alignment was always less than or equal to the length so this
wasn't seen before. Instead just apply device sector size rounding up
to the length.
Bug 756829 - SWRaid member detection enhancements
The super blocks for Linux Software RAID arrays using metadata types
0.90 and 1.0 are stored at the end of the partition and not currently
cleared by GParted.
Create a SWRaid array, stop it and format it to cleared using GParted.
The signature remains.
# mdadm --create /dev/md1 --level=linear --raid-devices=1 --force --metadata=1.0 /dev/sdb1
# mdadm --stop /dev/md1
[In GParted format to cleared /dev/sdb1]
# blkid /dev/sdb1
/dev/sdb1: UUID="8ac947a7-063f-2266-5f2a-e5d178198139" UUID_SUB="49bd51d4-4c54-fb16-a45e-bd795f783f59" LABEL="rockover:1" TYPE="linux_raid_member"
As fixed in other cases before [1][2] it is necessary to clear all
signatures before formatting as a new file system to prevent recognition
issues. For example now format the partition as a FAT32 file system.
Now there are two signatures and libparted reports one type and blkid
reports another.
# mkdosfs -F 32 /dev/sdb1
# blkid /dev/sdb1
/dev/sdb1: UUID="8ac947a7-063f-2266-5f2a-e5d178198139" UUID_SUB="49bd51d4-4c54-fb16-a45e-bd795f783f59" LABEL="rockover:1" TYPE="linux_raid_member"
# parted /dev/sdb print
Model: ATA SAMSUNG SSD UM41 (scsi)
Disk /dev/sdb: 8012MB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
1 1049kB 538MB 537MB primary fat32
(Deliberately avoided btrfs, ext2/3/4 and xfs as recent versions of
their mkfs tools clear other signatures first for the same reason).
[1] Bug 688882 - Improve clearing of file system signatures
[2] 3c75f3f5b1
Use wipefs to clear old signatures before creating new file systems (#688882)
Update erase_filesystem_signatures() to also zero the necessary sectors
to clear SWRaid metadata 0.90 and 1.0 super blocks.
Bug 756829 - SWRaid member detection enhancements
If the installation is unusual / broken such that the mdadm command is
not found but there are active SWRaid arrays, provide a fallback with
some of the information. In this case populate the cache with details
only available from /proc/mdstat. Information will be limited to active
arrays only and their members. No UUIDs or labels. There will be no
information about inactive arrays and GParted will use it's normal
libparted and blkid identification for those devices and partitions.
As mdadm has gained the capability to manage Fake/BIOS RAID arrays they
also appear in /proc/mdstat when mdadm is used to start them. Enhance
the parser of /proc/mdstat to only extract information for SWRaid arrays
with recognised metadata versions.
Bug 756829 - SWRaid member detection enhancements
Automatically load the cache of SWRaid information for the first time if
any of the querying methods are called before the first explicit
load_cache() call. Means we can't accidentally use the class and
incorrectly find no SWRaid members when they do exist.
Bug 756829 - SWRaid member detection enhancements
Busy file systems are accessed via a mount point, LVM Physical Volumes
are activated via the Volume Group name and busy SWRaid members are
accessed via the array device, /dev entry. Therefore choose to show the
array device in the mount point field for busy SWRaid members.
The kernel device name for an SWRaid array (without leading "/dev/") is
the same as used in /proc/mdstat and /proc/partitions. Therefore the
array device (with leading "/dev/") displayed in GParted will match
between the mount point for busy SWRaid members and the array itself as
used in the device combo box.
# cat /proc/mdstat
Personalities : [raid1]
md1 : active raid1 sda1[2] sdb1[3]
524224 blocks super 1.0 [2/2] [UU]
...
# cat /proc/partitions
major minor #blocks name
8 0 33554432 sda
8 1 524288 sda1
...
8 16 33554432 sdb
8 17 524288 sdb1
...
9 1 524224 md1
...
Bug 756829 - SWRaid member detection enhancements
In cases where blkid wrongly reports a file system instead of an SWRaid
member (sometimes confused by metadata 0.90/1.0 mirror array or old
version not recognising SWRaid members), the UUID and label are
obviously wrong too. Therefore have to use the UUID and label returned
by the mdadm query command and never anything reported by blkid or any
file system specific command.
Example of blkid reporting the wrong type, UUID and label for /dev/sda1
and the correct values for /dev/sdb1:
# blkid | egrep 'sd[ab]1'
/dev/sda1: UUID="10ab5f7d-7d8a-4171-8b6a-5e973b402501" TYPE="ext4" LABEL="chimney-boot"
/dev/sdb1: UUID="15224a42-c25b-bcd9-15db-60004e5fe53a" UUID_SUB="0a095e45-9360-1b17-0ad1-1fe369e22b98" LABEL="chimney:1" TYPE="linux_raid_member"
# mdadm -E -s -v
ARRAY /dev/md/1 level=raid1 metadata=1.0 num-devices=2 UUID=15224a42:c25bbcd9:15db6000:4e5fe53a name=chimney:1
devices=/dev/sda1,/dev/sdb1
...
ARRAY /dev/md127 level=raid1 num-devices=2 UUID=8dc7483c:d74ee0a8:b6a8dc3c:a57e43f8
devices=/dev/sdb6,/dev/sda6
...
NOTES:
* In mdadm terminology the label is called the array name, hence name=
parameter for array md/1 in the above output.
* Metadata 0.90 arrays don't support naming, hence the missing name=
parameter for array md127 in the above output.
Bug 756829 - SWRaid member detection enhancements
Add active attribute to the cache of SWRaid members. Move parsing of
/proc/mdstat to discover busy SWRaid members into the cache loading
code. New parsing code is a little different because it is finding all
members of active arrays rather than determining if a specific member is
active.
Bug 756829 - SWRaid member detection enhancements
Detection of Linux SWRaid members currently fails in a number of cases:
1) Arrays which use metadata type 0.90 or 1.0 store the super block at
the end of the partition. So file system signatures in at least
linear and mirrored arrays occur at the same offsets in the
underlying partitions. As libparted only recognises file systems
this is what is detected, rather than an SWRaid member.
# mdadm -E -s -v
ARRAY /dev/md/1 level=raid1 metadata=1.0 num-devices=2 UUID=15224a42:c25bbcd9:15db6000:4e5fe53a name=chimney:1
devices=/dev/sda1,/dev/sdb1
...
# wipefs /dev/sda1
offset type
----------------------------------------------------------------
0x438 ext4 [filesystem]
LABEL: chimney-boot
UUID: 10ab5f7d-7d8a-4171-8b6a-5e973b402501
0x1fffe000 linux_raid_member [raid]
LABEL: chimney:1
UUID: 15224a42-c25b-bcd9-15db-60004e5fe53a
# parted /dev/sda print
Model: ATA VBOX HARDDISK (scsi)
Disk /dev/sda: 34.4GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
1 1049kB 538MB 537MB primary ext4 boot, raid
...
2) Again with metadata type 0.90 or 1.0 arrays blkid may report the
contained file system instead of an SWRaid member. Have a single
example of this configuration with a mirrored array containing the
/boot file system. Blkid reports one member as ext4 and the other as
SWRaid!
# blkid | egrep 'sd[ab]1'
/dev/sda1: UUID="10ab5f7d-7d8a-4171-8b6a-5e973b402501" TYPE="ext4" LABEL="chimney-boot"
/dev/sdb1: UUID="15224a42-c25b-bcd9-15db-60004e5fe53a" UUID_SUB="0a095e45-9360-1b17-0ad1-1fe369e22b98" LABEL="chimney:1" TYPE="linux_raid_member"
Bypassing the blkid cache gets the correct result.
# blkid -c /dev/null /dev/sda1
/dev/sda1: UUID="15224a42-c25b-bcd9-15db-60004e5fe53a" UUID_SUB="d0460f90-d11a-e80a-ee1c-3d104dae7e5d" LABEL="chimney:1" TYPE="linux_raid_member"
However this can't be used because if a user has a floppy configured
in the BIOS but no floppy attached, GParted will wait for minutes as
the kernel tries to access non-existent hardware on behalf of the
blkid query. See commit:
18f863151c
Fix long scan problem when BIOS floppy setting incorrect
3) Old versions of blkid don't recognise SWRaid members at all so always
report the file system when found. Occurs with blkid v1.0 on
RedHat / CentOS 5.
The only way I can see how to fix all these cases is to use the mdadm
command to query the configured arrays. Then use this information for
first choice when detecting partition content, making the order: SWRaid
members, libparted, blkid and internal.
GParted shell wrapper already creates temporary blank udev rules to
prevent Linux Software RAID arrays being automatically started when
GParted refreshes its device information[1]. However an administrator
could manually stop or start arrays or change their configuration
between refreshes so GParted must load this information every refresh.
On my desktop with 4 internal hard drives and 3 testing Linux Software
RAID arrays, running mdadm adds between 0.20 and 0.30 seconds to the
device refresh time.
[1] a255abf343
Prevent GParted starting stopped Linux Software RAID arrays (#709640)
Bug 756829 - SWRaid member detection enhancements
When multiple devices are named on the command line and (after sorting
and removing duplicates) the device following a non-existent or invalid
one is not checked for usability [1]. In most situations this isn't
noticed as the device gets skipped at the "Searching ... partitions"
step instead. However as seen in bug 755495 and commit [2]
checking usability matters.
For example (on CentOS 6.5) a large sector disk device can be edited
when it follows a non-existent or invalid device named on the command
line:
# modprobe scsi_debug dev_size_mb=128 sector_size=4096
# fgrep scsi_debug /sys/block/*/device/model
/sys/block/sdd/device/model:scsi_debug
# ./gpartedbin /dev/does-not-exist /dev/sdd
======================
libparted : 2.1
======================
Could not stat device /dev/does-not-exist - No such file or directory.
Device /dev/sdd has a logical sector size of 4096. Not all parts of GNU Parted support this at the moment, and the working code is HIGHLY EXPERIMENTAL.
/dev/sdd: unrecognised disk label
When erasing a device don't skip confirming the following device is
usable.
[1] Usable device as implemented by useable_device()
Must not have a large sector size when GParted is built with an old
version of libparted which doesn't support large sector sizes and
must be able to read the first sector.
[2] 362b2db331
Check disks named on the command line are safe to use too (#755495)
Bug 756434 - GParted dumps core when passing non-existent or invalid
device on the command line
A non-existent or invalid disk device named on the command line caused
two libparted dialogs to be displayed repeatedly on every refresh. This
was because the device was only removed from the 'device_paths' vector
when it wasn't usable [1]; not when it didn't exist or was invalid, when
the libparted ped_device_get() call failed. Fix this.
[1] Usable device as implemented by useable_device()
Must not have a large sector size when GParted is built with an old
version of libparted which doesn't support large sector sizes and
must be able to read the first sector.
Bug 756434 - GParted dumps core when passing non-existent or invalid
device on the command line
Naming a non-existent or invalid disk device on the command line causes
GParted to dump core. Non-existent device looks like this:
# ./gpartedbin /dev/does-not-exist
======================
libparted : 2.4
======================
Could not stat device /dev/does-not-exist - No such file or directory.
Could not stat device /dev/does-not-exist - No such file or directory.
Backtrace has 10 calls on stack:
10: /lib64/libparted.so.0(ped_assert+0x31) [0x7fcfd10b3e61]
9: /lib64/libparted.so.0(+0x3fdfc12a0c) [0x7fcfd10b4a0c]
8: /home/mike/bin/gpartedbin-0.23.0-master-63-g23b5ba4() [0x455028]
7: /home/mike/bin/gpartedbin-0.23.0-master-63-g23b5ba4() [0x455090]
6: /home/mike/bin/gpartedbin-0.23.0-master-63-g23b5ba4() [0x4550d5]
5: /home/mike/bin/gpartedbin-0.23.0-master-63-g23b5ba4() [0x46723f]
4: /usr/lib64/libglibmm-2.4.so.1() [0x3ff5834a8d]
3: /lib64/libglib-2.0.so.0() [0x3fe086a374]
2: /lib64/libpthread.so.0() [0x3fdf407a51]
1: /lib64/libc.so.6(clone+0x6d) [0x3fdf0e893d]
Assertion (dev != NULL) at device.c:227 in function ped_device_open() failed.
Aborted (core dumped)
And with an invalid device the output looks like this:
# ./gpartedbin /dev/zero
======================
libparted : 2.4
======================
The device /dev/zero is so small that it cannot possibly store a file system or partition table. Perhaps you selected the wrong device?
Error fsyncing/closing /dev/zero: Invalid argument
The device /dev/zero is so small that it cannot possibly store a file system or partition table. Perhaps you selected the wrong device?
Error fsyncing/closing /dev/zero: Invalid argument
Backtrace has 10 calls on stack:
...
[Same as above]
Bisected the cause to this commit from 2015-03-09 in GParted 0.22.0. It
claimed to make no functional change. That turned out not to be true.
51ac4d5648
Split get_device_and_disk() into two (#743181)
Fix by simply adding the missed if condition in get_device().
Bug 756434 - GParted dumps core when passing non-existent or invalid
device on the command line
For probed DMRaid devices (when not using libparted DMRaid support)
GParted waits up to 1 second for udev to have processed all events and
created the /dev entries after starting each DMRaid array. This was
added by this commit from 2009-09-02:
e7352a5000
Ensure /dev file system device entries created before adding device
Do the same for devices named on the command line too.
Order named disk devices so that they appear in the combo box in the
same order which they would when probed. Also remove duplicates so that
the same disk devices aren't scanned multiple times and appear
duplicated in the UI.
Try this; it used to take ages to load and looked weird:
# gparted /dev/sda /dev/sdb /dev/sda /dev/sdb /dev/sda /dev/sdb
Bug 755495 - GParted allowing partitioning of large sector devices
specified on the command line, when built with old
libparted which doesn't support it
When probing for disk devices GParted ensures that libparted is capable
of handling the sector size safely and that it is a real disk before it
is shown in the UI. However when disk devices are named on the command
line none of these checks are performed.
Libparted versions before v2.2 can only safely handle a sector size of
512 bytes. Therefore on old distributions with libparted < v2.2 GParted
allows unsafe editing of disk devices with larger sector sizes when they
are named on the command line. Known to affect these distributions:
RedHat/CentOS 5 (parted 1.8.1)
RedHat/CentOS 6 (parted 2.1)
For example (on CentOS 6.5) large sector disk device is ignored when
probing:
# modprobe scsi_debug dev_size_mb=128 sector_size=4096
# fgrep scsi_debug /sys/block/*/device/model
/sys/block/sdd/device/model:scsi_debug
# gparted
======================
libparted : 2.1
======================
Device /dev/sdd has a logical sector size of 4096. Not all parts of
GNU Parted support this at the moment, and the working code is
HIGHLY EXPERIMENTAL.
Ignoring device /dev/sdd with logical sector size of 4096 bytes.
GParted requires libparted version 2.2 or higher to support devices
with sector sizes larger than 512 bytes.
However when the device is named it is not ignored and can be edited:
# gparted /dev/sdd
======================
libparted : 2.1
======================
Device /dev/sdd has a logical sector size of 4096. Not all parts of
GNU Parted support this at the moment, and the working code is
HIGHLY EXPERIMENTAL.
/dev/sdd: unrecognised disk label
Apply the same validity checks to disk devices named on the command line
as to probed ones.
Bug 755495 - GParted allowing partitioning of large sector devices
specified on the command line, when built with old
libparted which doesn't support it
Abstract code checking sector size and ensuring the first sector of a
candidate disk device can be read into new
GParted_Core::useable_device() method.
Bug 755495 - GParted allowing partitioning of large sector devices
specified on the command line, when built with old
libparted which doesn't support it
I missed another case of 'index' may be used uninitialised warning in
OperationDelete::apply_to_visual(). Indent a code block within an if
clause so that the compiler can confirm that the 'index' local variable
isn't used uninitialised. Prevent this compiler warning:
OperationDelete.cc: In member function 'virtual void GParted::OperationDelete::apply_to_visual(std::vector<GParted::Partition, std::allocator<GParted::Partition> >&)':
OperationDelete.cc:34: warning: 'index' may be used uninitialized in this function
Bug 755214 - Refactor operation merging
GParted waits forever when attempting to set a FAT16/32 file system
label which contains prohibited characters [1][2]. This is because
mlabel asks a question and is waiting for input. Force cancelling the
operation doesn't work either as GParted sends signal 2 (interrupt i.e.
[Ctrl-C]) but mtools commands specifically ignores this and a number of
other signals. Have to kill mlabel with signal 9 (kill) to regain
control of GParted.
Mlabel command with prohibited characters in the label:
# export MTOOLS_SKIP_CHECK=1
# mlabel ::"MYLABEL/ " -i /dev/sdb10
Long file name "MYLABEL/ " contains illegal character(s).
a)utorename A)utorename-all r)ename R)ename-all
s)kip S)kip-all q)uit (aArRsSq):
Remove prohibited characters from FAT16/32 file systems labels when
creating and labelling them. Also upper case the label to meet label
requirements [1][2]. This silently corrects the label and the actual
label applied will be displayed when GParted refreshes after applying
the operation.
[1] Microsoft TechNet: Label
https://technet.microsoft.com/en-us/library/bb490925.aspx
[2] Replicated in Wikikedia: label (command)
https://en.wikipedia.org/wiki/Label_%28command%29
Bug 755608 - Labeling fat16/fat32 partitions hangs if certain characters
included in label
These member variables store no Operation class information and were
being used as local variables. Replace with local variables.
Also indent a code block within an if clause so that the compiler can
confirm that the new local variable isn't used uninitialised. Prevents
this compiler warning:
OperationResizeMove.cc: In member function 'void GParted::OperationResizeMove::apply_normal_to_visual(std::vector<GParted::Partition, std::allocator<GParted::Partition> >&)':
OperationResizeMove.cc:125: warning: 'index' may be used uninitialized in this function
Bug 755214 - Refactor operation merging
Since this commit earlier in the patchset the second optional parameter
of method Win_GParted::Add_Operation() is no longer used. Remove it.
Replace open coded merge of resize/move into create operation
(#755214)
Bug 755214 - Refactor operation merging
This is the equivalent case fixed in the earlier commit, but now using
copy/paste to create the second new partition rather than plain new.
Fix visually re-apply create operation in create-create-grow-first
sequence (#755214)
Start with an existing partition as a copy source. Then this sequence
of operations will cause the copy partition to disappear from the disk
graphic:
1) create new #1,
2) copy existing / paste into unallocated leaving space preceding,
3) resize new #1 larger.
There are two different types of copy operation. The first is copy into
unallocated space creating a new partition which needs treating the same
as create new operation. The second is copy into existing partition
which needs treating the same as the other operations which don't change
the boundaries of the partition. Fix apply_to_visual() accordingly.
Bug 755214 - Refactor operation merging
Move the code from OperationCreate::apply_to_visual() into new method
Operation::insert_new() in the parent class. This is in preparation for
the following commit.
Bug 755214 - Refactor operation merging
The apply_to_visual() method for the change UUID, format, label file
system and name partition operations duplicated identical code. This
code was just substituting the partition in the disk graphic vector with
the new partition recorded in the operation, as none of these operations
change the partition boundaries. Move this duplicate code into the
parent class in new method Operation::substitute_new().
Bug 755214 - Refactor operation merging
After previous commit "Replace open coded merge of resize/move into
create operation (#755214)" the second created partition would disappear
from the disk graphic in the following sequence: create new #1, create
new #2 leaving space preceding, resize #1 larger. The create new #2
operation still existed and was shown in the operation list. It was
just that it disappeared from the disk graphic.
Remember that when each operation is created it records the partition,
or the unallocated space, to which the operation is applied at the time
the operation is created in the partition_original member variable. In
the above sequence the resize #1 larger operation was merged back into
the create new #1 operation. When visually re-applying the create
new #1 operation to the disk graphic, it left a smaller unallocated
partition following it. This was smaller than the unallocated partition
recorded in the create new #2 operation, hence it failed to visually
re-apply to the disk graphic.
The insight to fix this is that it doesn't matter what size the
unallocated space was when the create new operation was constructed. It
only matters that the new partition to be created fits in the available
unallocated space currently in the disk graphic.
Bug 755214 - Refactor operation merging
This information is already documented in the existing comments
associated with the calls to merge_operations() and assignments to the
mergetype variables. The table just summaries the rules together in one
place.
Bug 755214 - Refactor operation merging
For the case of resizing/moving a new, not yet created partition,
activate_resize() open coded the merge operation. Again this code has
existed in GParted since before version 0.0.5 and the current code
history in Git.
Replace the necessary code so that an explicit merge_operations() call
is used instead; along with the other case of resizing/moving an
existing partition.
NOTES:
This commit changes the merge direction. The old coded merged forward
by removing the old create operation and adding a new create operation
with the new size. This was bad because with multiple pending create
operations, each merged resize operation reordered those create
operations. Then when the operations were applied the partitions were
created and therefore numbered in a different order to that shown in
disk graphic.
The new code merges backwards by updating the initial create operation
with the new size. This maintains the create operation order so that
when applied the partitions are numbered in the same order as shown in
the disk graphic.
Bug 755214 - Refactor operation merging
For the case of formatting a new, not yet created partition,
activate_format() open coded the merge operation. This code has existed
in GParted since before version 0.0.5 and the current code history in
Git.
Replace the necessary code so that an explicit merge_operations() call
is used instead; along with the other case of formatting an existing
partition.
Bug 755214 - Refactor operation merging
Creation of the various operations involved various implicit rules about
how the different types of operations were merged in different cases.
This was open coded in each ::activate_*() method. Abstract this into
new merge_operations() method and make the merging rules explicitly
specified.
NOTE:
The removal of operation type checking in the MERGE_LAST_WITH_ANY cases
is not a problem because all the Operation*::merge_operations() methods
ensure the operation types match as part of the merge attempt.
Bug 755214 - Refactor operation merging
Rename Win_GParted::Merge_Operations() to merge_two_operations(). To
reflect what it does and in preparation for further refactoring of the
code.
Be more strict on the validation of the first and second indexes. The
first operation must also be before the second operation in the
operation[] vector. (It is actually a programming bug if first and
second fail validation. However so far g_assert() is only being used to
validate pointers, which if wrong would likely cause the program to
eventually crash when dereferenced later. In this case a bug would
merely cause the incorrectly specified pair of operations to not be
merged).
Move validate_display_partition_ptr() declaration in the header file to
be in the same ordering as it's definition in the source file.
Bug 755214 - Refactor operation merging
Win_GParted::Merge_Operations() method was modifying the internals of
Operation* objects; in particular the partition_new member variable.
This is breaking data hiding and encapsulation tenant of object oriented
programming.
Implement exactly the same operation merge semantics, but hide the
manipulation of the internals of the Operation* objects within the
Operation* classes themselves.
Bug 755214 - Refactor operation merging
... before refactoring the code.
See the commit message from 2011-10-05 for details of what operations,
available at that time, on the same partition can be merged and in what
cases:
b10349ae37
Merge overlapping operations (#438573)
Bug 755214 - Refactor operation merging
As previous commit, display_partitions is now a Win_GParted member
variable so checking for the existence of an extended partition can be
localised where it is used.
Remove index_extended member variable and localise the same checking in
activate_new().
Now that display_partitions is a Win_GParted member variable and
therefore available throughout the class, since commit [1], calculation
of primary_count can be localised in max_amount_prim_reached() where it
is used.
Implements a FIXME and removes primary_count as a member variable.
[1] 545b75d957
Move vector of partition objects to a Win_GParted class member (#750168)
Perform a copy, reformat source and paste sequence in GParted. When the
source is a primary partition everything works as expected, with the
newly pasted partition reflecting the reformatted source. However when
the source is a logical partition GParted thinks it is pasting the
original source, rather than the reformatted source. The same is also
true for other file system manipulation operations: resize, file system
label and new UUID. It is just that reformatting the source to a
different file system type is the most obvious in the UI and causes the
most significantly wrong actions to be performed.
For example start with an ext4 logical partition, select it for copy,
format it to xfs and paste into a new partition. GParted thinks the
second operation will create a copy of an ext4 file system instead of
the xfs file system. When applied the operation details are:
Format /dev/sdd5 as xfs
+ calibrate /dev/sdd5
+ clear old file system signatures in /dev/sdd5
+ set partition type on /dev/sdd5
+ create new xfs file system
+ mkfs.xfs -f -L "" /dev/sdd5
Copy /dev/sdd5 to /dev/sdd (start at 131.00 MiB)
+ calibrate /dev/sdd5
+ check file system on /dev/sdd5 for errors and (if possible) fix them
+ e2fsck -f -y -v -C 0 /dev/sdd5
e2fsck: Subperblock invalid, trying backup blocks...
Resize inode not valid. Recreate? yes
...
/dev/sdd5: ***** FILE SYSTEM WAS MODIFIED *****
+ create empty partition
+ set partition type on /dev/sdd6
new partition type: ext4
+ copy file system of /dev/sdd5 to /dev/sdd6
using internal algorithm
...
GParted formatted sdd5 to xfs, but then the copy step ran e2fsck and
managed to resurrect the ext4 file system and then performed a block
copy of it to partition sdd6. The copy step should have ran xfs_repair
and used xfsdump | xfsrestore to copy the xfs file system. Afterwards
sdd5 contains both xfs and ext4 signatures.
# wipefs /dev/sdd5
offset type
----------------------------------------------------------------
0x438 ext4 [filesystem]
UUID: f0ed4247-76db-4d93-b3bc-c7da4a70f95e
0x0 xfs [filesystem]
UUID: 1ac8e7c3-0311-4c64-8e4a-b715a23ea0bd
This has been broken at least as far back as GParted 0.1.0.
Fix by simply refreshing the copy source partition object when it is a
logical partition too, as well as when it is a primary partition.
Bug 754827 - Copy, reformat source and paste a logical partition thinks
it's pasting the original file system
Command exit status is a 1 byte value between 0 and 255. [1][2] However
at the Unix API level the value is encoded as documented in the
waitpid(2) manual page. This is true for the Glib API too. [3] This is
why, for example, the comment in ext2::check_repair() reported receiving
undocumented exit status 256. It was actually receiving exit status 1
encoded as per the waitpid(2) method.
Add shell style exit status decoding [2] to execution of all external
commands. Return value from Utils::execute_command() and
FileSystem::execute_command() functions are now:
0 - 125 - Exit status from the command
126 - Error executing the command
127 - Command not found
128+N - Command terminated by signal N
255 - Unexpected waitpid(2) condition
Also adjust checking of the returned statuses as necessary.
[1] Advanced Bash-Scripting Guide: Appendix D. Exit Codes With Special
Meanings
http://www.linuxtopia.org/online_books/advanced_bash_scripting_guide/exitcodes.html
[2] Quote from the bash(1) manual page:
EXIT STATUS
... Exit statuses fall between 0 and 255, though as
explained below, the shell may use values above 125
specially. ...
... When a command terminates on a fatal signal N, bash uses
the value of 128+N as the exit status.
If a command is not found, the child process created to
execute it returns a status of 127. If a command is found
but is not executable, the return status is 126.
[3] Quote from the Glib Reference Manual, Spawning Processes section,
for function g_spawn_check_exit_status():
https://developer.gnome.org/glib/stable/glib-Spawning-Processes.html#g-spawn-check-exit-status
The g_spawn_sync() and g_child_watch_add() family of APIs return
an exit status for subprocesses encoded in a platform-specific
way. On Unix, this is guaranteed to be in the same format
waitpid() returns, ...
Bug 754684 - Updates to FileSystem:: and Utils::execute_command()
functions
Replace open coding of the creation of the operation details for the
mlabel command used to set the label and UUID with calls to
FileSystem::execute_command() which will do it all. This also results
in the commands getting a time and check mark displayed in the operation
details.
Bug 754684 - Updates to FileSystem:: and Utils::execute_command()
functions
There has been an undocumented rule that external commands displayed in
the operation details, as part of file system manipulations, only get a
time and check mark displayed when multiple commands are needed, and not
otherwise. (GParted checks whether all commands are successful or not
regardless of whether a check mark is displayed in the operation details
or not).
EXCEPTION 1: btrfs resize
Since the following commit [1] from 2013-02-22, GParted stopped
displaying the timing for the btrfs resize command in the operation
details. It being part of a multi-command sequence to perform the step.
This is because FileSystem::execute_command() since the commit can only
check the exit status for zero / non-zero while timing and checking the
command status but btrfs resize needs to consider some non-zero statuses
as successful.
[1] 52a2a9b00a
Reduce threading (#685740)
EXCEPTION 2: ext2/3/4 move and copy using e2image
When use of e2image was added [2] the single command steps were timed
and check.
[2] 86111fe12a
Use e2image to move/copy ext[234] file systems (#721516)
EXCEPTION 3: fat16/32 write label and UUID
Uses Utils::execute_command() rather than FileSystem::execute_command()
so can be separately changed. See the following commit for resolution
of the final commands not yet timed and check mark displayed.
CHANGE:
Lets make a simpler rule of always displaying the time and a check mark
for all external commands displayed in the operation details. However
this makes several of the other single command actions need special exit
status handling because zero success, non-zero failure is not correct
for every case. Specifically affects resizing of reiserfs and check
repair of ext2/3/4, fat16/32, jfs and reiserfs.
After this change all external commands run as file system actions must
follow one of these two patterns of using the EXEC_CHECK_STATUS flag or
separately calling FileSystem::set_status() to register success or
failure of the command:
exit_status = execute_command(cmd, od, EXEC_CHECK_STATUS...);
or:
exit_status = execute_command(cmd, od, ...);
bool success = (exit_status == 0 || exit_status == OTHER_SUCCESS_VALUE...);
set_status(od, success );
Bug 754684 - Updates to FileSystem:: and Utils::execute_command()
functions
Change the two optional boolean parameters into a single optional flags
parameter which uses symbolically defined names. Makes reading the
execute_command() calls much easier to understand. (Implemented as bit
field using the same technique as used for Glib::SpawnFlags [1]).
This changes the calls thus:
execute_command(cmd, od) -> (cmd, od)
execute_command(cmd, od, false) -> (cmd, od, EXEC_NONE) // [2]
execute_command(cmd, od, true ) -> (cmd, od, EXEC_CHECK_STATUS)
execute_command(cmd, od, false, true) -> (cmd, od, EXEC_CANCEL_SAFE)
execute_command(cmd, od, true , true) ->
(cmd, od, EXEC_CHECK_STATUS|EXEC_CANCEL_SAFE)
[1] SpawnFlags bitwise operators in
/usr/include/glibmm-2.4/glibmm/spawn.h.
[2] False and EXEC_NONE are the default values for the optional third
parameter before and after this change respectively and both mean
the same. This is being used in btrfs::resize() and being kept for
now despite it being the default.
Bug 754684 - Updates to FileSystem:: and Utils::execute_command()
functions
Add a pattern to recognise Non-Volatile Memory Express devices as valid
devices to work with. Devices are named by the Linux kernel device
driver like /dev/nvme0n1 [1] with partitions named like /dev/nvme0n1p1
[2].
[1] linux 3.3 drivers/block/nvme.c nvme_alloc_ns()
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/block/nvme.c?id=v3.3#n1351
[2] Contents of /proc/partitions for a partitioned NVME device
$ grep nvme /proc/partitions
259 0 390711384 nvme0n1
259 1 977 nvme0n1p1
259 2 31250000 nvme0n1p2
259 3 328209496 nvme0n1p3
259 4 31249408 nvme0n1p4
Bug 755022 - gparted doesn't recognize nvme devices
The ntfs-3g package previously provided the FUSE based NTFS driver to
mount the file system and ntfsprogs provided the user space tools. In
April 2011 the packages have merged [1] forming ntfs-3g_ntfsprogs. Arch
Linux / Debian / Slackware / Ubuntu now just have an ntfs-3g package
with everything; where as CentOS / Fedora / openSUSE are sticking with
the original two package names. Reverse the order of the needed
packages to:
ntfs-3g / ntfsprogs
[1] Release: NTFS-3G + NTFSPROGS 2011.4.12
http://www.tuxera.com/release-ntfs-3g-ntfsprogs-2011-4-12/
For reiserfs the upstream package is named reiserfsprogs. Arch Linux /
Debian / Slackware / Ubuntu use this name; but CentOS / Fedora name it
reiserfs-utils and openSUSE names it just reiserfs. Update the README
file with all 3 package names.
reiserfsprogs / reiserfs-utils / reiserfs
In the File System Support dialog just use the first 2 package names as
we don't want to make the dialog too wide with all 3 names and hopefully
on openSUSE it is more obvious that the reiserfs package is needed to
support the reiserfs file system.
NOTE:
Again this slightly increases the width of the File System Support
dialog on my CentOS 6 desktop with default fonts, now from 676 to 707
pixels. Again still well within the 800 wide target and still narrower
than the main window.
Bug 753436 - Update documentation of GParted software dependencies
In Fedora bug report:
Bug 1176108 - Warning shown on BTRFS partition because of missing btrfs-tools package
https://bugzilla.redhat.com/show_bug.cgi?id=1176108#c0
The user said:
However there is no btrfs-tools package in the standard Fedora repo.
There is a btrfs-progs package, which is already installed. It's
unclear whether this is a real error or simply a mismatched package
name.
The upstream software is named btrfs-progs. Arch Linux / CentOS /
Fedora / Slackware use the upstream name. Debian / Ubuntu name it
btrfs-tools and openSUSE calls it btrfsprogs (no dash).
Rename the needed software to:
btrfs-progs / btrfs-tools
Upstream name first separated by slash from alternative names
distributions use.
NOTE:
This slightly increases the width of the File System Support dialog on
my CentOS 6 desktop with default fonts, from 655 to 676 pixels. Still
well within the 800 wide target and still narrower that the main window.
Bug 753436 - Update documentation of GParted software dependencies
This message would be displayed as part of the details for an operation
to create or format a partition as LVM2 PV on a disk using partition
table types dvh or pc98, which don't have an lvm flag.
v Format /dev/sdb1 as lvm2 pv
> calibrate /dev/sdb1
> clear old file system signatures in /dev/sdb1
> set partition type on /dev/sdb1
Skip setting unsupported partition flag: lvm
> create new lvm2 pv file system
Translators found this message difficult to translate, as reported in
bug 752901 - Add translator comment or fix string. It is also only
standard practice to set the lvm flag on partitions containing LVM2 PVs,
and not required. In hindsight reporting that an operation was skipped,
when the operation is not a necessity, is probably bad practice as it
could lead to uncertainty on behalf of the user and questions as to why
the flag isn't supported. Just remove the message.
Correct the colour name comment for NILFS2.
Colour for LVM2 PV was a Medium Brown. This is close to Face Skin Dark
from the GNOME colour palette. Use this instead.
Make colour for ReFS a bit darker so it is more distinct from the colour
for NTFS.
Add translation comments to improve the clarity of some translatable
text strings regarding setting partition flags.
Bug 752901 - Add translator comment or fix string
Opening the Resize/Move dialog on a logical partition causes GParted to
crash. This crash affects current GParted GIT HEAD, but does not affect
GParted 0.22.0. Git bisect identifies that it was broken with the
following commit:
Remove Set_Data() from the copy, resize/move and new dialog class APIs
7a4a375ed6
The problem was trying to treat the reference display_partitions_ref
like a pointer, and in particular on line 1732 trying to make it refer
to the a different vector of partitions, .logicals sub-vector.
1721 void Win_GParted::activate_resize()
1722 {
...
1726 std::vector<Partition> & display_partitions_ref = display_partitions;
1727 if ( selected_partition_ptr->type == TYPE_LOGICAL )
1728 {
1729 unsigned int ext = 0 ;
1730 while ( ext < display_partitions.size() && display_partitions[ext].type != TYPE_EXTENDED )
1731 ext++;
* 1732 display_partitions_ref = display_partitions[ext].logicals;
1733 }
1734
1735 Dialog_Partition_Resize_Move dialog( gparted_core.get_fs( selected_partition_ptr->filesystem ),
1736 *selected_partition_ptr,
1737 display_partitions_ref );
What was actually happening was that the .logicals sub-vector was being
copied, replacing the display_partitions vector and freeing the original
sub-vector. This left selected_partition_ptr pointing to the original
memory where the selected partition use to exist in the .logicals
sub-vector. At some point in the Dialog_Partition_Resize_Move class
*selected_partition_ptr was referenced, accessing the freed memory.
Crash soon followed.
Fix by using a pointer instead of a reference, which can be assigned to
point to a different object.
Bug 752587 - GParted crashing when opening Resize/Move dialog on
logical partition
Since GParted commit 52a2a9b "Reduce threading (#685740)", released in
GParted 0.15.0, application of operations occurs in the main thread
running the UI, therefore long running libparted actions such as
resizing a FAT16 or FAT32 file system hang the UI for as long as it take
to complete the operation.
https://git.gnome.org/browse/gparted/commit/?id=52a2a9b00a32996921ace055e71d0e09fb33c5fe
Though this problem exists for all libparted actions, it is particularly
noticeable when performing a large resize of fat16/fat32/hfs/hfs+ file
systems.
To address this significant cause of an unresponsive GUI, this
enhancement adds threading to the libparted ped_file_system_resize
function call.
Bug 737022 - UI hangs while running libparted operations such as
FAT16/FAT32 resizing
GParted was also searching the PATH for the availability of the udevadm
and udevsettle commands for every device with a busy partition during a
refresh and for every applied operation effecting a partition table. As
with hdparm previously this was wasteful.
Again, remember the result of searching the PATH at startup and refresh
when clicking on the [Rescan For Supported Actions] button in the File
System Support dialog.
Previously on every refresh for every device, GParted was searching the
PATH to discover if the hdparm command existed. Stracing GParted showed
that calling Glib::find_program_in_path("hdparm") made the following OS
calls:
access("/usr/lib64/qt-3.3/bin/hdparm", X_OK) = -1 ENOENT (No such file or directory)
access("/usr/local/sbin/hdparm", X_OK) = -1 ENOENT (No such file or directory)
access("/usr/local/bin/hdparm", X_OK) = -1 ENOENT (No such file or directory)
access("/sbin/hdparm", X_OK) = 0
getuid() = 0
stat("/sbin/hdparm", {st_mode=S_IFREG|0755, st_size=137, ...}) = 0
stat("/sbin/hdparm", {st_mode=S_IFREG|0755, st_size=137, ...}) = 0
The Linux VFS is very fast but repeatedly doing this is wasteful.
Remember the result of searching the PATH for the hdparm command at
startup and refresh this when the [Rescan For Supported Actions] button
is pressed in the File System Support dialog. This is the same as
GParted already does for file system specific commands and their
capabilities.
Bug 751251 - Show serial number in device information
With a Kobo Touch eReader connected via USB, GParted would hang when
scanning the device with this error written to the terminal:
$ sudo src/gpartedbin
======================
libparted : 2.3
======================
(gpartedbin:10261): glibmm-CRITICAL **:
unhandled exception (type Glib::Error) in signal handler:
domain: g_convert_error
code : 1
what : Invalid byte sequence in conversion input
The hdparm command was printing binary data as the serial number.
Fragment of the 'hdparm -I /dev/sdf' output:
# hdparm -I /dev/sdf
/dev/sdf:
SG_IO: bad/missing sense data, sb[]: 70 00 05 00 00 00 00 0a 00 00 00 00 24 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ATAPI Optical card reader/writer, with non-removable media
Model Number: {BINARY_DATA}
Serial Number: {BINARY_DATA}
Firmware Revision: {BINARY_DATA}
GParted reads command output using the Glib::IOChannel class. However
by default an IOChannel performs character set conversion on the data it
reads, so when it came across an invalid byte sequence in the binary
data the above exception was raised and the IOChannel::read() method
never returned. Hence GParted became stuck reportedly scanning the
same device forever. Code fragment:
src/PipeCapture.cc
49 bool PipeCapture::OnReadable( Glib::IOCondition condition )
50 {
...
58 Glib::ustring str;
>> 59 Glib::IOStatus status = channel->read( str, 512 );
60 if (status == Glib::IO_STATUS_NORMAL)
61 {
62 for( Glib::ustring::iterator s = str.begin(); s != str.end(); s++ )
Quote from the IOChannel class reference:
https://developer.gnome.org/glibmm/stable/classGlib_1_1IOChannel.html
Note that IOChannels implement an automatic implicit character set
conversion to the data stream, and usually will not pass by default
binary data unchanged. To set the encoding of the channel, use
e.g. set_encoding("ISO-8859-15"). To set the channel to no encoding,
use set_encoding() without any arguments.
Fix by disabling the automatic character set conversion in the IOChannel
used to read output from executed commands.
Bug 751251 - Show serial number in device information
Run "hdparm -I /dev/DISK" to get the hard drive serial number of
every device which has one and display it in the Device Information.
The displayed value can either be the actual serial number, "none" or
blank. "none" means the device doesn't have a hard drive serial number,
such as for Linux software RAID arrays, BIOS fake RAID arrays or USB
flash drives. Blank means something went wrong getting the serial
number. Either it couldn't be found in the hdparm output or the hdparm
command wasn't installed.
Example real hard drive:
# hdparm -I /dev/sda
...
ATA device, with non-removable media
Model Number: SAMSUNG HM500JI
Serial Number: S1WFJDSZ123732
...
Example Linux software RAID array:
# hdparm -I /dev/md127
/dev/md127:
HDIO_DRIVE_CMD(identify) failed: Inappropriate ioctl for device
On my desktop with 4 internal hard drives 2 Linux software RAID arrays
on those hard drives, 2 USB flash drives and 1 USB hard drive attached,
running hdparm 9 times added 0.07 seconds to the device refresh time.
Bug 751251 - Show serial number in device information
Btrfs-progs 4.1, released June 2015, includes support for changing the
UUID of a btrfs file system using the btrfstune command. Check for
availability by looking for the -u option in the btrfstune help output.
Use btrfstune like this:
# umount /dev/sdb1
# btrfstune -f -u /dev/sdb1
Current fsid: e7ad5dba-d721-4f99-990b-1ba2901c8ad2
New fsid: 231563d9-e173-410d-b1da-d34c4319a423
Set superblock flag CHANGING_FSID
Change fsid in extents
Change fsid on devices
Clear superblock flag CHANGING_FSID
Fsid change finished
# echo $?
0
Bug 751337 - btrfstune in btrfs-progs 4.1 supports changing the file
system UUID
Explain why the implementation uses a clear_cache() call, rather than a
straight load_cache() call. This commit from 2014-02-17 implemented
incremental loading of the btrfs device cache:
76e64f2905
Detect busy status of multi-device btrfs file systems (#723842)
The lvm query commands were always run and the cache loaded even if
GParted, actually blkid, didn't identify any LVM2 PVs. (GParted uses
libparted and blkid to identify partition content and the lvm commands
to provide the needed configuration details).
Now implement complete lazy initialization of the cache. Never force
loading of the cache. The cache is only loaded when the first value is
accessed from it. When there are no LVM2 PVs, the cache is never
queried, so never loaded. All the needed infrastructure for delayed
loading was previously added by this commit from 2011-12-11:
ff8ad04120
Lazy initialize the cache from querying LVM2 PVs (#160787)
Every public member function which access values from the cache already
calls initialize_if_required(). Just need to replace force loading of
the cache with a function which just clears the cache.
On my desktop, only when there are no LVM2 PVs, not loading the cache
and therefore not executing these external commands in
load_lvm2_pv_info_cache() saves 1.0 seconds of the 3.7 seconds it takes
to perform the a refresh in GParted:
lvm vgscan
lvm pvs ... -o pv_name,...
lvm pvs ... -o vg_name,...
Bug 750582 - Refactor the LVM2_PV_Info module object interface and
internal cache representation
GParted used to cache the results of the "lvm pvs" commands used to query
the state of the Logical Volume Manager as a series of lines of text.
Then every time a particular value was queried GParted would split all
the lines of text into fields until the required value was found.
Stop this repeat splitting of cached lines of text. Instead parse the
lines of text into separate fields and store in structures of values of
the correct type in the cache.
Bug 750582 - Refactor the LVM2_PV_Info module object interface and
internal cache representation
The LVM2_PV_Info cache had a pretend multi-object interface, yet all the
data is static. An LVM2_PV_Info object doesn't contain any member
variables, yet was needed just to call the member functions.
Make all the member functions static removing the need to use any
LVM2_PV_Info objects.
Bug 750582 - Refactor the LVM2_PV_Info module object interface and
internal cache representation
An LVM2_PV_Info object contains no member variables as all the data is
static (exists once in the program and accessed by all objects). The
constructor did nothing, except when passed true to load the cache.
Provide a separate load_cache() member function and remove the
constructors and destructor which do nothing. The C++ compiler will
provide a default constructor and destructor, which don't do anything as
there are no member variables to initialise and finalise.
This makes the interface a little easier to understand. Mostly a step
along the way of refactoring how the LVM2_PV_Info cache module works.
Bug 750582 - Refactor the LVM2_PV_Info module object interface and
internal cache representation
The copy, resize/move and new dialog classes (Dialog_Partition_Copy,
Dialog_Partition_Resize_Move and Dialog_Partition_New respectively) had
to be used like this:
construct dialog object passing some parameters
call Set_Data() to pass more parameters
run() dialog
call Get_New_Partition()
There is nothing in the classes which forces Set_Data() to be called,
but it must be called for the dialogs to work and prevent GParted from
crashing.
Make these class APIs safer by making it impossible to program
incorrectly in this regard. Move all the additional parameters from
each Set_Data() method to each constructor. The constructors just call
the now private set_data() methods.
The member variable was named selected_partition. It is assigned from
Win_GParted::selected_partition_ptr (which is a pointer to a const
partition object so is never updated). This gives connotations that it
won't be modified.
However it is updated freely as the new resultant partition object is
prepared before being returned from the dialog, most notable in the
Get_New_Partition() methods.
Therefore rename from selected_partition to new_partition.
The code goes like this:
Dialog_Partition_Copy::Get_New_Partition()
call Dialog_Base_Partition::Get_New_Partition()
Update this->selected_partition with results from running
the dialog.
return this->selected_partition by value.
Save value back to this->selected_partition.
Update this->selected_partition some more.
return this->selected_partition by value.
So there is an unnecessary copy of the partition object returned from
the base class Get_New_Partition() function back to the same variable in
the derived copy class Get_New_Partition() function.
Need to keep the base class Get_New_Partition() function as derived
class Dialog_Partition_Resize_Move uses that implementation as it
doesn't override it, and it's part of the interface.
Avoid this unnecessary copy by moving base class Get_New_Partition()
code into a new private function, called prepare_new_partition(), which
doesn't return anything. Then have Get_New_Partition() in both classes
just return the required partition object. Like this:
Dialog_Base_Partition::Get_New_Partition()
call prepare_new_partition()
return this->selected_partition by value.
Dialog_Partition_Copy::Get_New_Partition()
call Dialog_Base_Partition::prepare_new_partition()
Update this->selected_partition some more.
return this->selected_partition by value.
Bug 750168 - Reduce the amount of copying of partition objects
When Operation objects are created they take a copy of the Device object
to which the operation is to be applied. The Device object includes a
vector of all the contained Partition objects currently on the device,
so these get copied too.
These additional deep copied Partition objects in the Operation object
are never accessed. Therefore don't copy the contained Partition
objects when copying the Device object into the Operation object.
Bug 750168 - Reduce the amount of copying of partition objects
When opening the Manage Flags dialog, creation of the dialog object was
creating a copy of the selected partition object. If this was an
extended partition it also included recursively constructing the
contained logical partitions too.
Instead, replace the partition object in the DialogManageFlags class
with a reference to it.
Bug 750168 - Reduce the amount of copying of partition objects
When opening the Partition Information dialog, creation of the dialog
object was creating a copy of the partition object to be displayed. If
this was an extended partition it also included recursively constructing
the contained logical partitions too.
Instead, replace the partition object in the Dialog_Partition_Info class
with a reference to it.
NOTE:
In C++ a reference is really just a pointer under the hood. As such,
dereferences of a pointer to an object in the context of needing a
reference to the object doesn't copy the object. It merely initialises
the reference from the pointer.
Specifically, with this prototype:
Dialog_Partition_Info( const Partition & partition );
and the dialog object being constructed in Win_GParted::activate_info():
Dialog_Partition_Info dialog( *selected_partition_ptr );
the partition object is not copy constructed. A reference (pointer) to
it is merely passed to the dialog constructor.
Bug 750168 - Reduce the amount of copying of partition objects
The first actions of Win_GParted::activate_resize() were to create a
copy of the vector of partitions for the currently displayed device and
visually apply any pending operations. Exactly this has already been
done in Win_GParted::Refresh_Visual() with the result now available in
the member variable display_partitions. Stop this unnecessary partition
object copying and processing by just using display_partitions member
variable instead.
Bug 750168 - Reduce the amount of copying of partition objects
Document how GParted displays partitions in the GUI and manages the
lifetime and ownership of that data.
Bug 750168 - Reduce the amount of copying of partition objects
Further ensure that a bug doesn't get introduced with the use of
selected_partition_ptr, by asserting that it points to a current
partition object in the vector of display partitions.
After deliberately breaking the code so that selected_partition_ptr
points to some other partition object, trying to display the Information
dialog causes this crash:
======================
libparted : 2.4
======================
**
ERROR:Win_GParted.cc:989:void GParted::Win_GParted::set_valid_operations(): assertion failed: (valid_display_partition_ptr( selected_partition_ptr ))
Aborted (core dumped)
At this point in the code:
973 void Win_GParted::set_valid_operations()
974 {
...
986 // No partition selected ...
987 if ( ! selected_partition_ptr )
988 return ;
>> 989 g_assert( valid_display_partition_ptr( selected_partition_ptr ) ); // Bug: Not pointing at a valid display partition object
Bug 750168 - Reduce the amount of copying of partition objects
Add Glib g_assert() to ensure that a bug doesn't get introduced which
allows a partition callback to be called without a partition being
selected first.
After deliberately breaking the code so that selected_partition_ptr is
not set, trying to display the Information dialog causes this crash:
# ./gpartedbin
======================
libparted : 2.4
======================
ERROR:Win_GParted.cc:1978:void GParted::Win_GParted::activate_info(): assertion failed: (selected_partition_ptr != NULL)
Aborted (core dumped)
At this point in the code:
1976 void Win_GParted::activate_info()
1977 {
>> 1978 g_assert( selected_partition_ptr != NULL ); // Bug: Partition callback without a selected partition
1979
Bug 750168 - Reduce the amount of copying of partition objects
Now that TreeView_Details and DrawingAreaVisualDisk classes store and
pass pointers to partition objects in the Gtk signal callbacks, change
the selected partition into a pointer too.
Bug 750168 - Reduce the amount of copying of partition objects
This stops copying of each displayed partition object into the
DrawingAreaVisualDisk class.
Bug 750168 - Reduce the amount of copying of partition objects
This stops copying of each displayed partition object into the
TreeView_Details class.
It also stops copy constructing lots of partition objects when just
clicking on a partition in the disk graphic. The disk graphic needs to
inform the main GUI and then the partition list which partition has been
selected. The call sequence goes like:
DrawingAreaVisualDisk::on_button_press_event(event)
Win_GParted::on_partition_selected(partition_ptr, src_is_treeview)
TreeView_Detail::set_selected(partition_ptr)
TreeView_Detail::set_selected(rows, partition_ptr,
inside_extended)
Relevant source and highlighted comparison line:
140 bool TreeView_Detail::set_selected( Gtk::TreeModel::Children rows,
141 const Partition * partition_ptr, bool inside_extended )
142 {
143 for ( unsigned int t = 0 ; t < rows .size() ; t++ )
144 {
>> 145 if ( static_cast<Partition>( rows[t][treeview_detail_columns.partition] ) == *partition_ptr )
146 {
147 if ( inside_extended )
148 expand_all() ;
149
150 set_cursor( static_cast<Gtk::TreePath>( rows[ t ] ) ) ;
151 return true ;
152 }
153
154 if ( set_selected( rows[t].children(), partition_ptr, true ) )
155 return true ;
156 }
157
158 return false ;
159 }
Then in this function the partition selected in the disk graphic
(partition_ptr parameter) is compared in turn with each partition object
stored in the Gtk::TreeView model to find the matching one to mark it as
selected. This mere act of accessing the partition object stored in a
row of the Gtk::TreeView model causes it to be copy constructed. So
clicking on the 5th partition in the disk graphic will copy construct
the first 5 partition objects just to do a compare to find the matching
one.
This is because it is not possible to get a reference from a
Gtk:TreeViewProxy in gtkmm. Merely accessing a value in a Gtk::TreeView
model takes a copy of that value.
Subject: get a reference from a Gtk::TreeValueProxy
http://comments.gmane.org/gmane.comp.gnome.gtkmm/2217http://marc.info/?t=104400417500001&r=1&w=4
Bug 750168 - Reduce the amount of copying of partition objects
Change from passing a reference to the selected partition, to passing a
pointer to the selected partition in the signal_partition_selected
callbacks between the disk graphic, partition list and core GUI modules.
This is an enabler for the following patches.
Bug 750168 - Reduce the amount of copying of partition objects
Win_GParted::Refresh_Visual() used a local variable containing a copy of
the vector of partitions in the current device to be displayed. After
visually applying pending operations it loaded copies of each partition
object into the GUI widgets to display the disk graphic and partition
list, DrawingAreaVisualDisk and TreeView_Details classes respectively.
When a partition is selected in the UI, again a partition object is
copied. Also several of the partition dialogs, including the
information dialog, take a copy of the partition object. All these are
copies of the same set of partition objects, those currently being
displayed in the UI.
Move the vector of displayed partitions from a local variable in
Refresh_Visual() to a Win_GParted member variable. This will allow for
the above cases to be changed to used pointers and references to the
same set of partition objects.
The valid lifetime of pointers to elements in this partition object
vector is from one refresh to the next, when the vector is cleared and
repopulated with a new set of partition objects. This is exactly what
is needed as the GUI widgets are reloaded on each refresh, the selected
partition is reset and none of the partition dialog objects exist.
Dialog objects being created and destroyed on each use.
On the other hand some copies of partition objects currently being
displayed, still need to be made because they have lifetimes which need
to last longer than the next call to Refresh_Visual(). Specifically the
source of the copy partition and the partition objects copied into the
in the list of pending operations.
Bug 750168 - Reduce the amount of copying of partition objects
BUF in the copy dialog class, Dialog_Partition_Copy, is use to adjust
limits in 2 cases:
1) Minimum size when copying an XFS file system
Minimum size was set to the used space + 2 * cylinder size (typically
plus ~16 MiB). This commit from 2004-12-20 added it:
a54b52ea33
xfs copy now uses xfsdump and xfsrestore. icw some hacks in the other 2
Issues:
* This is increasing the minimum XFS file system size when copying it,
which doesn't happen in the resize case for other file systems.
* It allows an XFS file system to be created which is smaller than the
minimum size allowed by GParted. Copying an empty XFS file system can
create a new file system as small as 26 MiB. This is smaller than the
minimum GParted allows of 32 MiB because that is the minimum
xfs_repair can handle.
Remove this addition when copying an XFS file system and enforce minimum
file system size.
2) Maximum size when copying a file system into empty space larger than
it's maximum size
Maximum size was set to maximum file system size - cylinder size
(typically minus ~8 MiB). Only applied to FAT16 which has a maximum
file system size set in and can be grown. Added by this commit from
2004-12-15:
10e8f3338d
:get_fs now returns a const reference. in copy and resizedialog
...
* in copy and resizedialog filesystems with MAX set now have a max size of MAX - one cylinder .
Issue:
* This is applying a lower maximum resize when copying the file system
compared to that when creating the file system.
NOTE:
GParted currently allows all file systems to be resize to any size,
regardless of the maximum file system size. This is probably an
oversight, but it does allow libparted to convert FAT16 to FAT32 file
system when resizing.
Remove this lower maximum file system size when copying and resizing,
compared to creating.
Bug 749867 - Some limits are adjusted by arcane cylinder size amount
when copying and resizing in a single operation
This commit from 2010-05-20 removed use of cylinder size increase in the
minimum, and cylinder size decrease in the maximum file system sizes
from the resize/move dialog.
e62a23b5b5
Add partition alignment option to align to MiB (#617409)
This cylinder size limit adjustments were being performed using the
Dialog_Base_Partition::BUF member variable. Now in the
Dialog_Partition_Resize_Move class it is never accessed, and only
unnecessarily set. Move BUF from the common base class into the
Dialog_Partition_Copy class where it is still used.
Bug 749867 - Some limits are adjusted by arcane cylinder size amount
when copying and resizing in a single operation
Avoid long lines, long statements and repeated calls to
gparted_core.get_filesystem_object( selected_partition.filesystem ) by
storing the returned pointer in a local variable.
Needs the previous commit so that the the local variable can be a
pointer to a const FileSystem object instead of a pointer to a
(modifiable) FileSystem object.
The function never modifies any member variables so make it a const
member function.
(FileSystem::get_custom_text() is a virtual function so can't be made
static).
Rename a couple of GParted_Core methods for consistency and to better
distinguish get_filesystem() from get_filesystems() which do completely
unrelated things.
get_filesystem() -> detect_filesystem()
recognise_filesystem_signature() -> detect_filesystem_internal()
Also make detect_filesystem() a static member method as it doesn't use
any member variables. Requirement cascades to get_partition_path().
These member functions are only used within the GParted_Core class and
only operate on the static member variable FILESYSTEM_MAP.
Make both functions private and also make init_filesystems() static.
The FileSystem objects stored in the FILESYSTEM_MAP are allocated once
using new in init_filesystems() but never deleted.
Valgrind output fragment:
# valgrind --leak-check=full ./gparted
==29314== 353 (72 direct, 281 indirect) bytes in 1 blocks are definitely lost in loss record 6,287 of 6,905
==29314== at 0x4A075FC: operator new(unsigned long) (vg_replace_malloc.c:298)
>> ==29314== by 0x46EDA5: GParted::GParted_Core::init_filesystems() (GParted_Core.cc:106)
==29314== by 0x46EC5F: GParted::GParted_Core::GParted_Core() (GParted_Core.cc:96)
==29314== by 0x4A74F4: GParted::Win_GParted::Win_GParted(std::vector<Glib::ustring, std::allocator<Glib::ustring> > const&) (Win_GParted.cc:51)
==29314== by 0x4D600A: main (main.cc:56)
...
==29314== 161 (72 direct, 89 indirect) bytes in 1 blocks are definitely lost in loss record 6,119 of 6,905
==29314== at 0x4A075FC: operator new(unsigned long) (vg_replace_malloc.c:298)
>> ==29314== by 0x46F50C: GParted::GParted_Core::init_filesystems() (GParted_Core.cc:124)
==29314== by 0x46EC5F: GParted::GParted_Core::GParted_Core() (GParted_Core.cc:96)
==29314== by 0x4A74F4: GParted::Win_GParted::Win_GParted(std::vector<Glib::ustring, std::allocator<Glib::ustring> > const&) (Win_GParted.cc:51)
==29314== by 0x4D600A: main (main.cc:56)
GParted_Core.cc source:
102 void GParted_Core::init_filesystems()
103 {
104 FILESYSTEM_MAP[ FS_UNKNOWN ] = NULL ;
105 FILESYSTEM_MAP[ FS_CLEARED ] = NULL ;
>> 106 FILESYSTEM_MAP[ FS_BTRFS ] = new btrfs() ;
...
>> 124 FILESYSTEM_MAP[ FS_XFS ] = new xfs() ;
125 FILESYSTEM_MAP[ FS_BITLOCKER ] = NULL ;
Fix by deleting all FILESYSTEM_MAP pointers. Note that delete on a NULL
pointer is defined by C++ as a safe do nothing operation.
C++ FAQ / Do I need to check for null before delete p?
https://isocpp.org/wiki/faq/freestore-mgmt#delete-handles-null
Fixing this reduces the valgrind reported definitely lost memory blocks
count from 25 down to 6. 19 FileSystem objects deleted and 19 memory
blocks no longer lost.
Bug 749036 - FileSystem objects are memory leaked in init_filesystems()
Background:
GParted_Core::calibrate_partition() reloads the partition path name and
boundary to ensure they are correct before the operation is performed.
(See comments in calibrate_partition() for the reasons why this is
necessary). This also displays details of the partition being modified
in the operation details to inform the user.
The operation object contains these relevant member objects:
* partition_original
Partition before the operation is applied.
* partition_new
Partition as it is intended to be after the operation has been
applied.
* partition_copied (for the copy operation only)
Source partition being copied.
Issues:
GParted_Core::apply_operation_to_disk() was always calibrating partition
object partition_original, but for about half the operations
partition_original was not used and partition_new is used, so should be
calibrated instead.
Copy into an existing partition calibrated three partitions, the source,
destination before and destination after the operation was applied.
This doesn't really make sense in the operation details to the user.
They would expect to only see the source and destination partitions and
don't care about the distinction between the before and after
representation of the destination.
Minor issues:
The previous fix had to copy the correct partition path from the
calibrated partition_original object to the used partition_new object
for the format, label file system, name partition and change uuid
operations.
Calibrate was called for the create operation too, even though the
partition didn't yet exist. It was a no-operation.
Fix:
Stop always calibrating the partition_original object and instead
calibrate the correct partition object in each operation case. For the
copy into existing partition operation only calibrate the right two
partition objects as the user would expect.
Bug 746559 - Various operations fail when following paste into existing
partition
Format, label file system and new UUID operations would fail when
applied in a sequence to the destination partition following a previous
copy-paste operation.
Giving the copy of a file system a new label and a new UUID are the sort
of actions which should be performed when the disk containing the copy
remains attached to the same computer. This really should work.
Fragment of the failing operation details for a copy and label operation
sequence:
+ Copy /dev/sdb1 to /dev/sdb2
+ calibrate /dev/sdb2
+ calibrate copy of /dev/sdb1
+ calibrate /dev/sdb1
+ check the file system on /dev/sdb1 for errors and (if possible fix them
+ copy file system of /dev/sdb1 to /dev/sdb2
+ Set file system label "small-dst" on copy of /dev/sdb1
+ calibrate copy of /dev/sdb1
path: /dev/sdb2 (partition)
...
+ set file system label to "small-dst" on copy of /dev/sdb1
+ e2label copy of /dev/sdb1 "small-dst"
Usage: e2label device [newlabel]
This is failing because the file system specific command is passed
"copy of /dev/sdb1" as the device name. Code sequence:
1) OperationCopy::OperationCopy() sets the real path name of the
partition_new object to "copy of /dev/SRC" for display purposes.
2) GParted_Core::apply_operation_to_disk() calls calibrate_partition()
on partition_original object, restoring the real path name for
object partition_original.
3) apply_operation_to_disk() calls format(), label_filesystem() or
change_uuid() on the partition_new object, which still has the real
path name set to "copy of /dev/SRC". File system specific commands
fail with this as a path name.
Fix by copying the real path name from object partition_original to
partition_new, as is already done for the resize/move operation. Also
apply this fix to the name partition operation, because it uses the
partition_new object and so that it displays the real path name in the
operation details.
Bug 746559 - Various operations fail when following paste into existing
partition
When the partition is named in the Create New Partition dialog, set the
partition name as part of the create partition operation. Currently
this is only supported for GPTs. See
Utils::get_max_partition_name_length() for details.
Bug 746214 - Partition naming enhancements
Add a partition name entry box to the Create New Partition dialog. The
entry box is greyed out (not sensitive) for partition table types which
don't support partition naming. Currently only supported for GPTs. See
Utils::get_max_partition_name_length() for details.
There was a slightly wider gap between the file system combobox row and
the label entry row when there were only three widgets on the right hand
side of the dialog. This has been removed now that there are four
widgets so that they are all evenly spaced and they line up with the
four widgets on the left hand side.
So far the partition name can be entered and previewed, but isn't yet
applied to the disk.
Bug 746214 - Partition naming enhancements
Adding a partition name entry to the Create New Partition dialog will
need access to these two Device methods: partition_naming_supported()
and get_max_partition_length(). The Set_Data() function already takes
two parameters, only_unformatted and disktype, taken from Device member
variables.
Rather than add two more parameters to the Set_Data() function pass the
Device object instead, replacing the current only_unformatted and
disktype parameters.
Bug 746214 - Partition name enhancements
This is a small tidy-up to remove Gtk::Entry method calls on the file
system label entry box in the Create New Partition dialog which serve no
purpose.
filesystem_label_entry.set_activates_default( true );
It trying to make the Create New Partition dialog automatically
close when Enter is pressed with focus in the label entry box.
However this doesn't work, presumably because the default widget for
the dialog is not the Add button. Remove.
filesystem_label_entry.set_text( partition.get_filesystem_label() );
Initialises the text in the entry box with the file system label
from the passed partition object. The label is blank and the entry
box defaults to blank. Achieves nothing. Remove.
filesystem_label_entry.select_region( 0, filesystem_label_entry.get_text_length() );
Highlights the empty text in the entry box. Achieves nothing.
Remove.
NOTE:
The same set of Gtk::Entry method calls in Dialog_FileSystem_Label() and
Dialog_Partition_Name, which are editing the existing file system label
and partition name respectively, do work and have a useful effect so
shouldn't be removed.
Rename Gtk::Entry object entry -> filesystem_label_entry in the
Dialog_Partition_New class. This is in preparation for the introduction
of the partition name entry box in the Create New Partition dialog.
Bug 746214 - Partition name enhancements
Preview of the format operation cleared the partition name, yet when
applied, the partition name reappeared. Fix the preview to reflect
reality.
Bug 746214 - Partition naming enhancements
Previously partition naming had only been implemented for gpt. Make the
code ready to support naming of the other partition table types for
which libparted supports naming. Specifically: amiga, dvh, mac and
pc98 in addition to gpt. Document issues found with some of these
partition table types, which can relatively easily been worked around.
Leave support of naming for partition table types other than gpt
disabled, mostly just to reduce ongoing testing effort, at least until
there is any user demand for it.
Bug 746214 - Partition naming enhancements
Allow partition names to be changed whether or not the partition is
busy, rather than only when not busy, because it doesn't effect the busy
file system or change the partition boundaries in any way.
Bug 746214 - Partition naming enhancements
Attempting to create a new partition on a pc98 partition table fails
with the following libparted error:
The flag 'lvm' is not available for pc98 disk labels.
This has been broken since LVM2 Physical Volume read-write support was
first added in this commit:
c3ab62591b
Add creation of LVM2 PVs (#670171)
Fix by only clearing and setting the lvm partition flag when the type of
the partition table supports it. When creating a partition to contain
an LVM2 PV and the lvm flag is not support add the following message to
the operation results to explain that setting the lvm partition flag was
skipped and why:
Skip setting unsupported partition flag: lvm
Bug 746204 - Creating partitions on pc98 table fails with lvm flag not
available
Refactor GParted_Core::set_partition_type().
1) Set lp_partition variable earlier and use a single if lp_partition
set condition, rather than in both if conditions for the normal file
system case and the LVM2 Physical Volume case.
2) Stop calling Utils::get_filesystem_string() multiple times, instead
save the result in a local variable.
Tidies the code a little and reorders it in preparation for the
following fix to only set the lvm partition flag when support, making
that code change simpler.
Bug 746204 - Creating partitions on pc98 table fails with lvm flag not
available
resize_move() and move() stopped using the device parameter in this
commit from 2006-07-23:
d663c3c277
removed cylindersize buffering during resize from the filesystems. It is
create() stopped using the device parameter in this commit from 2006-03-19:
ad9f2126e7
fixed issues with copying (see also #335004) cleanups + added FIXME added
For reference most other operation methods had the device parameter
removed in this earlier commit from 2005-12-07:
642f0a145b
from now on each partition has a reference to it's device. make use of new
When the following conditions were met GParted would fail to recognise a
newly created whole disk device file system, and instead show an unknown
file system filling the disk:
1) Disk was previously partitioned and contained at least one partition.
2) Using libparted version 2.0 to 3.0 inclusive.
Initial status:
# blkid | fgrep sdc
# fgrep sdc /proc/partitions
8 32 976762584 sdc
8 33 104857600 sdc1
# parted /dev/sdc
GNU Parted 2.4
Using /dev/sdc
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) print
Model: ATA ST1000LM024 HN-M (scsi)
Disk /dev/sdc: 1000GB
Sector size (logical/physical): 512B/4096B
Partition Table: msdos
Number Start End Size Type File system Flags
1 1049kB 107GB 107GB primary
When creating the loop partition table libparted would not inform the
kernel to delete the old partitions. /proc/partitions still contained
the details of the old partitions.
(parted) mktable loop
Warning: The existing disk label on /dev/sdc will be destroyed and
all data on this disk will be lost. Do you want to continue?
Yes/No? Yes
(parted) print
Model: ATA ST1000LM024 HN-M (scsi)
Disk /dev/sdc: 1000GB
Sector size (logical/physical): 512B/4096B
Partition Table: loop
Number Start End Size File system Flags
(parted) quit
# fgrep sdc /proc/partitions
8 32 976762584 sdc
8 33 104857600 sdc1
Creation of the whole disk device file system goes unnoticed by blkid
because the kernel and therefore blkid's cache have stale partition
information.
# mkfs.xfs -f /dev/sdc
# blkid | fgrep sdc
NOTE:
On a Linux Software RAID array, as opposed to a hard disk, blkid does
notice creation of the whole disk device file system. However the
kernel still has old partition details.
This was fixed in libparted 3.1 by commit:
http://git.savannah.gnu.org/cgit/parted.git/commit/?id=f5c909c0cd50ed52a48dae6d35907dc08b137e88
libparted: remove has_partitions check to allow loopback partitions
Fix by deleting old partitions before creating the loop table when
compiled with a broken version of libparted. The GParted UI provides
no feedback while a new partition table is created, and with some
versions of GTK the UI become unresponsive too, so it is important to be
as fast as possible. Evaluated three different methods, deleting 15 and
22 MSDOS partitions on a physical 5400 RPM hard drive using libparted
2.4:
M1) Delete and commit one partition at a time.
Takes up to 24 seconds to delete 15 partitions. With 22 partitions
libparted always reports finding some of the partitions busy and
unable to inform the kernel about the modifications.
Too slow and doesn't work.
M2) Delete all partitions in one go and commit once.
Takes up to 1.4 seconds to delete either 15 or 22 partitions. Never
removes partitions 17 and higher from the kernel.
Doesn't work.
M3) Write GPT table (letting libparted delete any old partitions).
Takes up to 0.8 seconds to delete either 15 or 22 partitions.
Fast and works.
Use method 3 - write a GPT table thus using libparted code to inform the
kernel of the old partition deletions.
Bug 743181 - Add unpartitioned drive read-write support
Older versions of blkid don't correctly distinguish between FAT16 and
FAT32 file systems when overwriting one with the other. This effects
GParted too with these file systems on whole disk devices where only
blkid is used to recognise the contents. See previous fix for why only
blkid is used in this case:
Avoid whole disk FAT being detected as MSDOS partition table
(#743181)
Example:
# blkid -v
blkid from util-linux 2.20.1 (liblkid 2.20.0, 19-Oct-2011)
# mkdosfs -F16 -I /dev/md1
# blkid | fgrep md1
/dev/md1: SEC_TYPE="msdos" UUID="7C23-95D9" TYPE="vfat"
# mkdosfs -F32 -I /dev/md1
# blkid | fgrep md1
/dev/md1: SEC_TYPE="msdos" UUID="7F93-98F4" TYPE="vfat"
So blkid recognised the UUID changed but didn't remove the SEC_TYPE for
the FAT32 file system. See FS_Info::get_fs_type() as it uses this to
distinguish between FAT16 and FAT32. This is a caching update bug in
blkid, because telling blkid not to use the cache gets the right
results:
# blkid -c /dev/null | fgrep md1
/dev/md1: UUID="7F93-98F4" TYPE="vfat"
With testing determined that blkid from util-linux 2.23 and later are
not affected and earlier versions are affected. Mostly recently known
affected distribution is Ubuntu 14.04 LTS with util-linux 2.20.1.
The straight forward fix would be to instruct blkid to not use its cache
with 'blkid -c /dev/null'. But using blkid's cache is needed to prevent
blkid hanging for minutes when trying to access a non-existent floppy
drive when the BIOS is set incorrectly. See commit:
18f863151c
Fix long scan problem when BIOS floppy setting incorrect
Instead, when using an older affected version of blkid and when blkid
cache reports a vfat file system, run blkid again bypassing the cache.
The device is known to exist and contain a vfat file system, just not
whether it is a FAT16 or FAT32 file system, so can't be a non-existent
floppy device and won't hang.
Bug 743181 - Add unpartitioned drive read-write support
Libparted 1.9.0 to 2.3 inclusive, recognises whole disk device FAT file
systems as MSDOS partition tables. This causes GParted to do the same.
# dd if=/dev/zero bs=1M of=/dev/md4
# mkdosfs -F32 -v -I /dev/md4
# blkid /dev/md4
/dev/md4: UUID="53FE-31F2" TYPE="vfat"
# parted /dev/md4
GNU Parted 2.1
Using /dev/md4
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) print
Model: Unknown (unknown)
Disk /dev/md4: 536MB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
(parted) quit
# /tmp/parted24/bin/parted /dev/md4
GNU Parted 2.4
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) print
Model: Linux Software RAID Array (md)
Disk /dev/md4: 536MB
Sector size (logical/physical): 512B/512B
Partition Table: loop
Number Start End Size File system Flags
1 0.00B 536MB 536MB fat32
(parted) quit
This was fixed in libparted 2.4 by commit:
http://git.savannah.gnu.org/cgit/parted.git/commit/?id=616a2a1659d89ff90f9834016a451da8722df509
libparted: avoid regression when processing a whole-disk FAT partition
Make GParted immune to this bug by moving blkid performed whole disk
device file system detection before libparted partition detection. Also
have to always erase old file system signatures on whole disk devices
when creating new partition tables to ensure that blkid doesn't detect
those old signatures before libparted has a chance to detect the new
partition table.
Bug 743181 - Add unpartitioned drive read-write support
When writing "loop" partition table over the top of some whole disk
device file system types GParted continued to show those whole disk
device file systems rather than the virtual unknown partition from the
"loop" partition table.
This affected btrfs, jfs, reiser4 and reiserfs. It occurred because of
several factors:
1) Libparted only zeroed the first and last 9.5 KiB (assuming 512 byte
sectors) of the device before writing a new partition table. See
ped_disk_clobber().
2) These file systems have their super blocks and therefore signatures
after the first 9.5 KiB.
3) Whole disk device file system detection is performed using blkid
before checking for a libparted "loop" partition table. See
GParted_Core::set_devices_thread().
Ref:
libparted 3.2: disk.c:ped_disk_clobber()
http://git.savannah.gnu.org/cgit/parted.git/tree/libparted/disk.c?id=v3.2#n302
Fix by always erasing any possible file system signatures on the device
before creating a new "loop" partition table.
NOTE:
This is typically taking up to 0.5 seconds in my testing on a 5400 RPM
hard drive, during which time the GParted UI is hung and the create
partition table dialog shows the apply button pressed but no other
progress indication.
Bug 743181 - Add unpartitioned drive read-write support
Creating a new partition table was getting libparted to read any
existing partition table before creating a new partition table on the
device. This is an unnecessary step, and if the device didn't already
contain a partition table also printed this error from libparted:
/dev/sdb: unrecognised disk label
Since get_device_and_disk() has been split into two, just call
get_device() instead to just populate the PedDevice object representing
the disk device. Removes a small unnecessary step.
Bug 743181 - Add unpartitioned drive read-write support
The preview of clearing a whole disk device file system was previewing
the same as formatting to all other file system types; as a cleared file
system spanning the whole disk device. However when implemented this
removes all signatures on the disk so it actually becomes an unallocated
and unpartitioned device. Make the preview match what happens in when
implemented.
GParted previously used mydevice.max_prims = -1 to represent an
unpartitioned device. It is now represented as:
mydevice.max_prims = 1
mydevice.disktype = _("unrecognized")
mydevice.partitions[0].type = TYPE_UNALLOCATED
mydevice.partitions[0].whole_device = true
mydevice.partitions[0].filesystem = FS_UNALLOCATED
and the check for an unpartitioned device in Win_GParted.cc becomes:
partitions[0].type == TYPE_UNALLOCATED && partitions[0].whole_device
Bug 743181 - Add unpartitioned drive read-write support
Previously GParted displayed a device containing the parted "loop"
partition table signature "GNU Parted Loopback 0" and nothing else, as
an unrecognised device.
Now make GParted display this as a virtual whole disk device partition
with unknown contents, complete with the unable to detect a file system
warning. This change then allows a whole disk device file system to be
created with the following two steps:
1) Create "loop" partition table on a device;
2) Format to required file system.
GParted represents a whole disk device file system as:
mydevice.max_prims = 1
mydevice.disktype = "none"
mydevice.partitions[0].type = TYPE_PRIMARY
mydevice.partitions[0].whole_device = true
mydevice.partitions[0].filesystem = FS_EXT4 (example)
Now represents just Parted's "loop" signature as:
mydevice.max_prims = 1
mydevice.disktype = "loop"
mydevice.partitions[0].type = TYPE_PRIMARY
mydevice.partitions[0].whole_device = true
mydevice.partitions[0].filesystem = FS_UNKNOWN
And as before, an unpartitioned device as:
mydevice.max_prims = -1
mydevice.disktype = _("unrecognized")
mydevice.partitions[0].type = TYPE_UNALLOCATED
mydevice.partitions[0].whole_device = true
mydevice.partitions[0].filesystem = FS_UNALLOCATED
Bug 743181 - Add unpartitioned drive read-write support
Only allow resizing, not moving of a whole disk device file system.
There is no actual partition to move and moving a file system away from
the start of a disk only makes it unrecognisable.
Also don't perform the partition resize step as there's no actual
partition to be resized. Only the file system is being resized.
(Libparted actually allows the virtual partition spanning a whole disk
device to be resized, implementing it as a no-operation, but only for
recognised file systems. For unrecognised file systems it fails with
"unrecognised disk label").
Note that the existing resize dialog was designed for resizing partition
boundaries, and their contained file systems, not for resizing file
systems within a fixed boundary. The difference is noticeable when
there is unallocated space because the file system doesn't fill the
whole disk device. The dialog starts resizing a virtual partition the
size of the whole disk device, not the actual size of the file system.
Leave addressing this for a possible future update.
Bug 743181 - Add unpartitioned drive read-write support
First, copying into a whole disk device fails on the set partition type
step. Fails with either libparted error "The flag 'lvm' is not
available for loop disk labels" or "unrecognised disk label" depending
whether libparted recognised the content and created a virtual partition
or not. (This is with libparted 2.4).
Fix by just skipping setting the partition type on whole disk devices.
Second, if any file system specific tools are used during the copy, they
will fail because they are passed the device name as "copy of /dev/SRC"
instead of "/dev/DST". Occurs when either the destination whole disk
device is not an identical size to the source so the file system check
and grow steps are added, or when file system specific tools are used to
copy the file system as with XFS or recent EXT2/3/4 tools.
Fix by re-adding the real partition path from libparted for whole disk
devices, as is already done for partitioned device names in
GParted_Core::calibrate_partition().
Bug 743181 - Add unpartitioned drive read-write support
Creation of reiserfs file system fails in GParted with the this error.
# mkreiserfs -f --label "" /dev/sdb < /dev/null
mkreiserfs 3.6.24
/dev/sdb is entire device, not just one partition!
Continue (y/n):
# echo $?
1
Add second force flag, -f, to the mkreiserfs command to make it work.
Bug 743181 - Add unpartitioned drive read-write support
Creation of ext2/3/4 and ntfs file systems fails in GParted on whole
disk devices with these errors.
# mkfs.ext4 -L "" /dev/sdb < /dev/null
mke2fs 1.42.9 (4-Feb-2014)
/dev/sdb is entire device, not just one partition!
Proceed anyway? (y,n)
# echo $?
1
# mkntfs -Q -v -L "" /dev/sdc
/dev/sdc is entire device, not just one partition.
Refusing to make a filesystem here!
# echo $?
1
Add force flag, -F, to both mkfs commands to make them work.
Bug 683643 - Doesn't properly support partitionless drives.
Formatting a whole disk device fails on the set partition type step with
libparted error "unrecognised disk label". This is because the previous
step just cleared the old file system signatures leaving libparted with
nothing to recognise. Therefore libparted doesn't present a virtual
"loop" partition table.
As there is no partition table, there's no partition and no partition
type. Just skip setting the partition type on whole disk devices.
Bug 743181 - Add unpartitioned drive read-write support
This enables Format to Cleared operation to succeed on whole disk device
file systems even when libparted doesn't recognise the file system.
(Turns out that making calibrate work in the previous commit happened to
make Format to Cleared operation succeed, but only if libparted
recognised the file system on the whole disk device).
Bug 743181 - Add unpartitioned drive read-write support
In the operational results of the calibrate step include the type of the
path GParted is working with, either partition or whole disk device. Do
the same for the create empty partition step too for consistency, even
though it only ever creates partitions. Looks like:
create empty partition
path: /dev/sdb3 (partition)
start: 2099200
end: 4196351
size: 2097152 (1.00 GiB)
calibrate /dev/sdc
path: /dev/sdc (device)
start: 0
end: 1953525167
size: 1953525168 (931.51 GiB)
Makes it explicit to the users what GParted has detected. Helps the
developers when looking at saved results to understand what decisions
were made and why specific steps were performed or not.
Bug 743181 - Add unpartitioned drive read-write support
This enables the Check, Label and New UUID operations to succeed on
whole disk device file systems even when libparted doesn't recognise the
file system.
This benefits reiser4 and lvm2 pv file systems with all versions of
libparted, current version is 3.2, and for nilfs2 with libparted < 2.4.
Bug 743181 - Add unpartitioned drive read-write support
get_device_and_disk() basically calls libparted to get a PedDevice
object representing a disk device and a PedDisk object representing a
partition table. Re-implement get_device_and_disk() using two separate
functions, get_device() and get_disk(), to get one of these objects
each.
No functionality changes with this commit. It enables future commits to
incrementally add support for whole disk devices into GParted without
needing libparted to recognise the contents and create a virtual "loop"
partition table.
Bug 743181 - Add unpartitioned drive read-write support
Enable operations on whole disk devices containing any recognised file
system.
The new partition operation on an empty whole disk device continues to
display the "No partition table found on device /dev/DEVICE" information
dialog.
Specifically unsupported operations:
* Delete -
Deletion of a partition only involves removal of the entry in the
partition table leaving the file system intact on the disk. However
this doesn't work for a whole disk device file system. Instead the
file system signatures would have to be erased which is much more
destructive and virtually impossible to undo. Therefore don't
implement whole disk device file system deletion. Alternatives are
to format the file system to cleared or create a partition table on
the device. Both of these imply overwriting the existing data and
set the expectation that undo is not possible.
* Manage flags -
There's no partition table, so there's no partition, so there's no
flags.
Resize/Move operation is being supported so that a whole disk device
file system can be resized to handle devices which can be resize, such
as those from SANs or Linux Software RAID arrays. The start of the file
system must remain fixed so move won't be allowed.
So far only simple operations work if they don't need libparted support
at all [1], or only need libparted support for the calibrate step AND
the file system on the whole disk device is recognised by libparted [2].
(Needs libparted to provide a "loop" partition, hence the recognition
requirement, so that the calibrate step can successfully read the
virtual "loop" partition table. Doesn't matter whether it's an old
version of libparted and it gets the name of the device wrong as GParted
is already using the whole disk device name anyway).
[1] Operations not needing any libparted support:
Mount on, Unmount, Swapon, Swapoff, Activate and Deactivate
[2] Operations only needing libparted support for the calibrate step:
Check, Label, New UUID
Bug 743181 - Add unpartitioned drive read-write support
Need to be able to take different actions in the GParted_Core partition
manipulation methods and in Win_GParted UI methods to deal with
libparted supported partitions or whole disk devices without a partition
table. Add boolean whole_device to the partition object and set
appropriately to allow for this.
Bug 743181 - Add unpartitioned drive read-write support
As was done with a failed mount operation, include the failed activate/
deactivate command in the error dialog. Two example error dialogs now
look like this:
(-) Could not deactivate swap
# swapoff -v /dev/sdb7
swapoff: /dev/sdb7: swapoff failed: Invalid argument
[ OK ]
(-) Could not unmount /dev/sdb6
# umount -v "/mnt/6"
umount: /mnt/6: not mounted
[ OK ]
On RHEL/CentOS 6, GParted fails to mount nilfs2 file system like this:
# mkfs.nilfs2 /dev/sdb1
# mount /dev/sdb1 /mnt/1
mount: you must specify the filesystem type
This fails because mount internally uses libblkid to determine the file
system type when it is not specified on the command line. However on
RHEL/CentOS 6 libblkid is too old to recognise nilfs2.
GParted used libparted recognition first and blkid second. Mount only
uses libblkid. When there are multiple signatures on a partition
GParted may report a different result to blkid alone. Therefore fix by
first trying to mount the file system without specifying the type, as is
already done, and if that fails, trying specifying the file system type.
This allows GParted to mount nilfs2 file systems.
# mount -t nilfs2 /dev/sdb1 /mnt/1
# mount | fgrep sdb1
/dev/sdb1 on /mnt/1 type nilfs2 (rw,gcpid=30946)
And for unsupported file systems the error dialog from the failed mount
command shows both commands like this:
(-) Could not mount /dev/sdb3 on /mnt/3
# mount -v /dev/sdb3 "/mnt/3"
mount: unknown filesystem type 'reiser4'
# mount -v -t reiser4 /dev/sdb3 "/mnt/3"
mount: unknown filesystem type 'reiser4'
[ OK ]
Bug 742741 - Nilfs2 file system is unusable on RHEL/CentOS 6
Function Utils::get_filesystem_kernel_name() returns the name of the
file system as needed for use in the mount command:
mount -t TYPE DEVICE DIR
Needed because the kernel / mount name is 'hfsplus' where as libparted /
GParted, as reported by Utils::get_filesystem_string(), calls it 'hfs+'.
So far just added debugging when mounting a file system to test the
function works.
# ./gartedbin
======================
libparted : 2.1
======================
DEBUG: (hfsplus) # mount -v /dev/sdb5 "/mnt/5"
DEBUG: (nilfs2) # mount -v /dev/sdb1 "/mnt/1"
Bug 742741 - Nilfs2 file system is unusable on RHEL/CentOS 6
On RHEL/CentOS 6 GParted fails to mount nilfs2 file system. Include the
failing file system mount command in the error dialog so the user knows
what command failed. The error dialog now looks like:
(-) Could not mount /dev/sdc1 on /mnt/1
# mount -v /dev/sdc1 "/mnt/1"
mount: you must specify the filesystem type
[ OK ]
Also stop telling the dialog that the secondary text contains pango
markup as the command line and error message certainly isn't pango
markup text.
Bug 742741 - Nilfs2 file system is unusable on RHEL/CentOS 6
GParted can create a nilfs2 file system on RHEL/CentOS 6, yet both
libparted and blkid are too old to recognise it when created. Add
GParted internal detection of nilfs2 file systems.
Easy way to get the required signature needed to recognise nilfs2 is
using wipefs to report the signature bytes it erases, using a more up to
date distribution. So using CentOS 7:
# mkfs.nilfs2 /dev/sdb1
# wipefs -a /dev/sdb1
/dev/sdb1: 2 bytes were erased at offset 0x00000406 (nilfs2): 34 34
Also from util-linux source libblkid/src/superblocks/nilfs2.c
#define NILFS_SB_MAGIC 0x3434
Bug 742741 - Nilfs2 file system is unusable on RHEL/CentOS 6
For libparted recognised file systems covering the whole disk device,
libparted reports the partition table as "loop". With libparted 3.1 and
earlier, the partition device was reported wrongly as /dev/sdb1, instead
of /dev/sdb, by the ped_partition_get_path() call. This causes GParted
to fail to report file system usage, label and UUID, and busy status.
Also multiple file system tool command errors from using the wrong
device are displayed.
# mkfs.ext4 -L whole-ext4 /dev/sdb
# ./gpartedbin
Example partition errors:
e2label: No such file or directory while trying to open /dev/sdb1
Couldn't find valid filesystem superblock.
tune2fs 1.42.8 (20-Jun-2013)
tune2fs: No such file or directory while trying to open /dev/sdb1
Couldn't find valid filesystem superblock.
dumpe2fs 1.42.8 (20-Jun-2013)
dumpe2fs: No such file or directory while trying to open /dev/sdb1
Couldn't find valid filesystem superblock.
Unable to read the contents of this file system!
Because of this some operations may be unavailable.
This cause might be a missing software package.
The following list of software packages is required for ext4 file
system support: e2fsprogs v1.41+.
This also breaks most of the manipulation operations, again because it
is using the wrong device name.
Fix by ignoring libparted's "loop" partition table and just let it get
displayed via the "none" partition table mechanism.
Doing this renders the direction taken in bug 683643 to make GParted
work with libparted "loop" partition tables mute, as GParted no longer
works with such partition tables. Instead they are replaced by "none"
partition tables within GParted.
Bug 683643 - Doesn't properly support partitionless drives
https://bugzilla.gnome.org/show_bug.cgi?id=683643
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
So that, even though the menu item is greyed out, it is correctly named
Activate or Deactivate for an LVM2 Physical Volume on a whole disk
device, rather than the default Mount or Unmount.
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
At this time we are providing read-only support for file systems on
whole disk devices. Also all manipulation operations fail at the
calibrate step with libparted error "unrecognised disk label", and
manage flags is for real partitions only.
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
For file systems which libparted recognises, when found on the whole
disk device, it reports with partition table "loop" and a partition
covering the whole disk. GParted duly displays this to the user.
For file systems which libparted doesn't recognise it reports
"unrecognised disk label". As of the latest libparted 3.2, these file
system aren't recognised and can't currently be shown when on the whole
disk device:
BitLocker, Crypt LUKS, exFAT, F2FS, LVM2 Physical Volume,
Linux Software RAID, ReFS, Reiser 4
So only when libparted doesn't recognise a file system on the whole disk
device and GParted does, either via blkid or it's internal code, display
this with partition table "none".
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
Move code which queries the file system label and UUID of a partition
into a separate helper function.
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
Make GParted_Core::get_filesystem() capable of recognising file system
signatures on whole disk devices. Pass NULL for lp_partition to make it
read signatures from the whole disk rather than a specific partition.
Add debugging to test it works.
# fgrep sdb /proc/partitions
8 16 8388608 sdb
# lvm pvcreate /dev/sdb
Writing physical volume data to disk "/dev/sdb"
Physical volume "/dev/sdb" successfully created
# ./gpartedbin
======================
libparted : 1.8.1
======================
Unable to open /dev/sdb - unrecognised disk label.
DEBUG: /dev/sdb (lvm2 pv)
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
Used 'fstype' as the name of local variable storing the FILESYSTEM
enumeration type.
'fs_type' was the name of the local variable storing the string with the
name of the file system type. Rename it to 'fsname' to distinguish it
from the aforementioned variable.
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
Refactor GParted internal file system signature detection to remove code
duplication. There were 5 separate copies of code to: allocate a
buffer, open, read and close the device, free the buffer and compare the
signature.
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
(1) Add recognition of LUKS signatures detected by blkid command, via
the FS_Info cache.
(2) Relocate internal GParted detection of LUKS signature to back after
libparted and blkid identification. This reverses previous commit:
7695593d21
Improve crypt-luks detection (#578300)
Bug 578300 must have been caused by either: (1) reading stale data from
the whole disk device, or (2) formatting as LUKS over the top of ext3
leaving old signatures behind. Case (1) has been resolved by commit:
3bea067596
Flush devices when scanning to prevent reading stale signatures
(#723842)
Re-tested case (2) and find that no old signatures are left behind:
# mkfs.ext3 /dev/sda8
# wipefs /dev/sda8
offset type
----------------------------------------------------------------
0x438 ext3 [filesystem]
UUID: 756dcd5f-73c5-4881-81ee-dc88e8b18f42
# cryptsetup luksFormat /dev/sda8
# wipefs /dev/sda8
offset type
----------------------------------------------------------------
0x0 crypto_LUKS [crypto]
UUID: de15ed1a-f461-46b7-803b-d0fd3b4bd680
Bug 741430 - GParted cannot recognise LVM signature on unpartitioned
drive
Trying to create an XFS file system causes the following error and
the create new xfs file system step to never complete.
# ./gpartedbin
======================
libparted : 2.4
======================
(gpartedbin:15572): glibmm-CRITICAL **:
unhandled exception (type Glib::Error) in signal handler:
domain: g-shell-error-quark
code : 0
what : Text ended before matching quote was found for ". (The text was 'mkfs.xfs -f -L "mylabel" " /dev/sdb8')
Creation of XFS file systems was broken by the extra stray double quote
added to the command before the device name, by commit:
63aeb150ac
Rename member variables and methods in Partition class (#741424)
Bug 744108 - Creating an XFS file system never completes
Encountered this dialog:
Libparted Bug Found!
(-) Invalid argument during seek for read on /dev/md1
[ Retry ] [ Cancel ] [ Ignore ]
After investigation found that GParted was asking libparted to read from
sector -1. Libparted was actually reporting an error but GParted was
incorrectly displaying this as a bug. This was because setting the
dialog title was falling through from the error case to the bug case.
Fix this.
Also provide unique titles for all libparted exception types so that
they can all be distinguished.
* List of exceptions from libparted: enum _PedExceptionType
http://www.gnu.org/software/parted/api/group__PedException.html
Bug 743825 - gparted displaying dialog: libparted bug found!
Provide credit for patch to add support for GPT partition names.
Also indicate significant effort by Sinlu Bes while we tried different
ideas for how GPT partition name support should work.
Embedded devices (Android) use GPT partition names to identify
partitions, instead of file system labels. Add support for viewing and
changing them.
As partition names are used to provide unique identification they are
never copied when copying the contents of one partition to another.
Note that GNU/Linux uses file system labels, UUIDs or device names for
identification during the boot process and afterwards so while partition
names can be used, they are optional and purely for user information.
Bug 741424 - Add support for GPT partition names
This and the following few commits rename variables, methods, classes,
etc from *label_partition* to *label_filesystem* so that the code also
reflects that it is the label of the file system that is being modified
and to separate it from the name partition operation about to be added.
enum OPERATION_LABEL_PARTITION -> OPERATION_LABEL_FILESYSTEM
Bug 741424 - Add support for GPT partition names
Rename the Partition Label concept to File System Label throughout the
GUI. This is to better reflect that it is the label of the file system
that is being modified and to separate it from the Partition Name
concept about to be added.
Bug 741424 - Add support for GPT partition names
GParted reports used figure of active swap space from 4 KiB upwards,
actually 1 page, where as 'swapon -s' reports figures from 0 upwards.
This is because GParted is counting the 1 page swap space overhead as
used space.
# sfdisk -s /dev/sdb1
262144
# mkswap /dev/sdb1
# swapon /dev/sdb1
# swapon -s
Filename Type Size Used Priority
/dev/sdb1 partition 262140 0 -1
For this case GParted reports used as 4 KiB.
(ptn size - swap size = 262144 - 262140 = 4)
Instead make GParted report used figure without any overhead from 0
upwards, matching what 'swapon -s' reports.
Bug 742647 - Swap usage doesn't match figures reported by swapon -s
GParted's internal block copy has an off by one sector bug when the
source is before the destination; and the copy is performed backwards
from high block to low block. It is as though the source and
destination partitions were both one sector earlier on the disk.
In ASCII art it it looks like this:
Initial layout: x<--SRC--><--DST-->
Actually wrote: x<--SRC--
Should have written: <--SRC-->
Affects moving partitions too.
This bug has existed since commit:
bd9e16f22f
Thread the internal copy algorithm (#685740)
Effectively the last sector of the partition is missed and one sector
before the start of the partition is corrupted. Most of the time file
systems don't get around to using the very last sector so didn't notice.
Bug 742920 - GParted internal copy is overstepping partition boundaries
With reiser4progs 1.0.7 and earlier mkfs specifically checks for Linux
kernel 2.5 or 2.6. On Linux 3.x it fails with the following warning:
# mkfs.reiser4 --yes --label "" /dev/sdb1
Warn : Linux 3.13.0-39-generic is detected. Reiser4 does not support
such a platform. Use -f to force over.
# echo $?
8
This is fixed in reiser4progs 1.0.8 and later. From the reiser4progs
SourceForce README.txt:
. Do not ask for confirmation to work under Linux-3.X;
Found on Ubuntu 12.04 LTS and Ubuntu 14.04 LTS both with reiser4progs
1.0.7 and a Linux 3.x kernel.
Fix by adding "--force" flag to the mkfs command line.
Bug 742511 - Creation of reiser4 file systems fails with unsupported
warning
Split the resizing file system and resizing partition calls in
GParted_Core::resize() into separate grow and shrink code paths.
Note that this also changes the degenerative case of calling resize()
when the partition isn't changing size, for non-swap partitions, from a
file system check step to a successful no-op. This doesn't matter as my
testing never found resize() to be called when the partition isn't
changing size.
Also correct spelling of local variable success.
Bug 741211 - Remove unnecessary duplicate actions when resizing a
partition
Shrinking swap partition operation performs these steps:
Shrink /dev/sdb2 from 2.00 GiB to 1.00 GiB
+ calibrate /dev/sdb2
+ check file system on /dev/sdb2 for errors and (if possible) fix them
checking is not available for this file system
+ shrink file system
+ create new linux-swap file system
+ mkswap -L "test-swap" -U "bd381eba-5df7-42e2-8e0e-411e9701c995" /dev/sdb2
+ shrink partition from 2.00 GiB to 1.00 GiB
+ create new linux-swap file system
+ mkswap -L "test-swap" -U "bd381eba-5df7-42e2-8e0e-411e9701c995" /dev/sdb2
Resizing a linux-swap partition was partially special cased in
GParted_Core::resize(). Make it fully special cased so that it just
does the following steps. No more skipped file system checks or extra
resizing.
1) Resize partition,
2) Recreate linux-swap.
This existing call chain recreates the linux-swap:
GParted_Core::resize_filesystem()
linux_swap::resize()
A compound move and resize operation still performs unnecessary checks
and recreates of linux-swap, but less than before.
Bug 741211 - Remove unnecessary duplicate actions when resizing a
partition
Shrinking a partition performs a single file system check step, but
growing a partition still performs two file system checks.
Grow /dev/sda8 from 1.00 GiB to 10.00 GiB
+ calibrate /dev/sda8
+ check file system on /dev/sda8 for errors and (if possible) fix them
+ grow partition from 1.00 GiB to 10.00 GiB
+ check file system on /dev/sda8 for errors and (if possible) fix them
+ grow file system to fill the partition
This is a leftover from the early days of GParted when resizing a
partition used cylinder buffering and performed three file system
checks. See prior to commit:
d663c3c277
removed cylindersize buffering during resize from the filesystems.
Remove the second file system check when growing a partition.
Bug 741211 - Remove unnecessary duplicate actions when resizing a
partition
Simplify how the list of file system types is populated in the Create
New Partition dialog. Change from copying everything and removing
unwanted items to only copying required items. Makes the code simpler
and therefore easier to understand.
Only recognises ReFS file system. No other actions are supported.
Requires blkid from util-linux >= 2.24.
Bug #738471 - ReFS file system is not recognised
Use GParted_Core::supported_filesystem() to remove the need to
explicitly list the growing number of recognised, but otherwise
unsupported, file system signatures in multiple places.
Bug #738471 - ReFS file system is not recognised
Helper to check whether a recognised file system type is supported by
GParted or not. Supported means there is an implementation class and
will appear in the File System Support dialog.
Make supported_filesystem() a static member function so that it can be
called without a class object so that GParted_Core::GParted_Core()
initialiser isn't called multiple times. This requires FILESYSTEM_MAP
to become a static member variable too.
Bug #738471 - ReFS file system is not recognised
get_filesystem_object() takes a constant reference to a FILESYSTEM, but
FILESYSTEM is just an enumeration. So that's a pointer to a constant
int. Just pass by value instead.
RHEL / CentOS 5.6 and later officially support ext4 file system [1].
From RHEL / CentOS 5.3 ext4 file system was included as a technology
preview. Ext4 file system tools are in a separate package e4fsprogs,
using uniquely named commands. The standard e2fsprogs commands only
support ext2 and ext3 file systems.
# mkfs.ext4 /dev/sdb3
# tune2fs -l /dev/sdb3
tune2fs 1.39 (29-May-2006)
tune2fs: Filesystem has unsupported feature(s) while trying to open /dev/sdb3
Couldn't find valid filesystem superblock.
# echo $?
1
# tune4fs -l /dev/sdb3
tune4fs 1.41.12 (17-May-2010)
Filesystem volume name: <none>
Last mounted on: <not available>
Filesystem UUID: ba4a9d58-7728-4b47-8a90-80e772615637
Filesystem magic number: 0xEF53
Filesystem revision #: 1 (dynamic)
Filesystem features: has_journal ext_attr resize_inode dir_index filetype extent flex_bg sparse_super large_file huge_file uninit_bg dir_nlink extra_isize
...
For ext4 only, search for the e4fsprogs specific commands first and the
standard e2fsprogs commands second.
[1] RHEL 5.6 Release Notes, 5. Filesystems and Storage
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/5/html/5.6_Release_Notes/ar01s05.html
Bug #738706 - GParted doesn't support ext4 on RHEL/CentOS 5.x
Automatic selection of the largest unallocated partition caused flashing
redraw of the partition graphic and partition list. Both the partition
graphic and partition list were being drawn blank then redrawn fully
populated. This only happened on some distributions including:
CentOS 5.10, 6.5, 7.0, Debian 6, Fedora 14, 20, Xubuntu 14.04 LTS. Did
not happen on: Kubuntu 12.04 LTS.
This is a workaround, not a complete fix. It moves automatic selection
of the largest unallocated partition to after processing of the GTK
Event loop in Refresh_Visual() which redraws the partition graphic and
partition list. These visuals are now drawn only once, fully populated,
however this draws them without the selected partition. The partition
selection is then drawn afterwards which causes the selection to flash
instead. This is significant improvement to the whole partition graphic
and partition list flashing when redrawn.
Bug #696149 - Double refresh of display introduced with default
unallocated space
The device combobox was getting drawn blank, then getting drawn again
with the selected device. This was happening because at the start of
Win_GParted::refresh_combo_devices() the GTK model behind the combobox,
liststore_devices, was cleared, changing the active item, causing the
combobox to get redrawn empty. After the GTK model had been repopulated
the active item was reset causing the comboxbox to get redrawn again,
now showing the selected device. Call flow:
Win_GParted::refresh_combo_devices()
liststore_devices->clear()
//Gtk::Combobox emits signal_change. Registered callbacks
//called.
Win_GParted::combo_devices_changed()
Win_GParted::Refresh_Visual()
...
...
combo_devices.set_active(current_device);
//Gtk::Combobox emits signal_change. Registered callbacks
//called.
Win_GParted::combo_devices_changed()
Win_GParted::Refresh_Visual()
...
This has always been the case, since the device combobox was first added
to GParted before version 0.1 by commit:
3a4b43e0ad
replaced deprecated OptionMenu with ComboBox ...
Fix by temporarily blocking the devices comboxbox from emitting
signal_changed while the GTK model behind the combobox is recreated.
However, since automatic selection of the largest free space was added
[1] in GParted 0.15.0, a more noticeable flashing redraw issue was
caused in which the partition graphic and partition list were both drawn
blank then redrawn fully populated. Some distributions were not
affected by this at all, some only experienced a single flash and others
suffered from two or more flashing redraws. Some affected
distributions: CentOS 5.10, 6.5, 7.0, Debian 6, Fedora 14, 19, 20,
Ubuntu 13.10, Xubuntu 14.04 LTS. Did not occur on Kubuntu 12.04 LTS.
[1] 5b53c12f6e
Select largest unallocated partition by default (#667365)
Bug #696149 - Double refresh of display introduced with default
unallocated space
Automake 1.13 and later generates the following warning when building
GParted from git:
$ ./autogen.sh
...
checking for automake >= 1.9...
testing automake-1.13... found 1.13.4
...
Running automake-1.13...
src/Makefile.am:1: warning: 'INCLUDES' is the old name for 'AM_CPPFLAGS' (or '*_CPPFLAGS')
...
INCLUDES name has been depreciated since automake 1.7. GParted's
autogen.sh and configure.ac require at least automake 1.9. Therefore
replace automake variable INCLUDES with AM_CPPFLAGS.
Bug #735742 - automake warns INCLUDES is deprecated
Use AC_ARG_ENABLE() and AS_HELP_STRING() macros to improve handling of
the --disable-doc option.
Change the names, and sence, of the Autoconf and Automake definitions
used to control building of the help documentation. Reasons are:
(1) Switch from negative to positive statements which are more natural
to comprehend.
(2) Insert HELP in the names from *_DOC to *_HELP_DOC as they control
building of the documentation in the help subdirectory, not the
manual page in the doc directory.
Autoconf: HAVE_DISABLE_DOC -> ENABLE_HELP_DOC
Automake: DISABLE_DOC -> BUILD_HELP_DOC
Bug #734076 - Autodetect parted online partition resizing capability
Currently -lparted is passed to the linker by being hard coded in the
gpartedbin_LDFLAGS variable in src/Makefile.am and -lparted-fs-resize
by conditionally being appended to the gpartedbin_LDFLAGS.
Add an Autoconf AC_CHECK_LIB check for the parted library using it's
default action-if-found to extend LIBS. Update the existing
AC_CHECK_LIB check for the parted-fs-resize library so that the
overridden action-if-found also extends LIBS with -lparted-fs-resize.
These libraries are then assigned via the LIBS variable in the
generated Makefiles.
This matches how the other libraries, dl and uuid, were found in the
configure script and passed to the Makefiles.
Bug #734718 - Update Autoconf version specific libparted checks and
defines to feature specific ones
Remove HAVE_LIBPARTED_2_2_0_PLUS definition
Rename version specific #define HAVE_LIBPARTED_2_2_0_PLUS into two
separate feature specific names set as required:
Name Set when?
ENABLE_PT_REREAD_WORKAROUND (libparted < 2.2)
USE_LIBPARTED_LARGE_SECTOR_SUPPORT (libparted >= 2.2)
Using feature specific #defines is the standard Autoconf way and makes
the resultant conditional code easier to understand. Still have to
check the version of libparted though.
Bug #734718 - Update Autoconf version specific libparted checks and
defines to feature specific ones
Remove HAVE_LIBPARTED_3_0_0_PLUS definition
The #define only controls whether fat16 and fat32 file systems are moved
by libparted (for versions <= 2.4) or the GParted internal algorithm
(for libparted versions >= 3.0). See this commit for more details:
0fda1d011d
Enable new fs resize library available with parted-3.1 (#668281)
As nearly all other file system are moved using the GParted internal
algorithm, just always use it to move fat16 and fat32 file systems too.
Bug #734718 - Update Autoconf version specific libparted checks and
defines to feature specific ones
This patch changes the reading of the btrfs multi-device membership to
resolve issue 1/2 by ignoring the exit status from the 'btrfs filesystem
show' command and relying on parsing the required information to
determine success or failure.
Bug #733601 - Btrfs: Warnings and missing label with btrfs-progs 3.12
and 3.14
Patch 3/4 - btrfs::read_label()
This patch changes the btrfs file system usage reading code to resolve
issue 1/2 by ignoring the exit status from the 'btrfs filesystem show'
command and relying on parsing the required information to determine
success or failure.
Bug #733601 - Btrfs: Warnings and missing label with btrfs-progs 3.12
and 3.14
Issue 2/2 - GParted doesn't show label for mounted btrfs file systems
'btrfs filesystem show /dev/PTN' command is used to query details of a
btrfs file system including reading the file system label. When the
file system is mounted the label is no longer enclosed in single quotes,
but only when using btrfs-progs v3.12. This causes GParted to think the
label is blank when the file system is mounted and therefore no longer
display it.
File system label not enclosed in single quotes when mounted:
# fgrep sdb1 /proc/mounts
/dev/sdb1 /mnt/1 btrfs rw,relatime,space_cache 0 0
# btrfs filesystem show /dev/sdb1
Label: test1-btrfs uuid: 1f78fa38-2f85-41d3-9be6-ae0356ae9469
Total devices 1 FS bytes used 192.00KiB
devid 1 size 2.00GiB used 240.75MiB path /dev/sdb1
Btrfs v3.12
File system label enclosed in single quotes when unmounted:
# umount /dev/sdb1
# btrfs filesystem show /dev/sdb1
Label: 'test1-btrfs' uuid: 1f78fa38-2f85-41d3-9be6-ae0356ae9469
Total devices 1 FS bytes used 192.00KiB
devid 1 size 2.00GiB used 240.75MiB path /dev/sdb1
Btrfs v3.12
Removing the single quotes enclosing the label makes the output
identical to that from the older 'btrfs-show' command.
Fix by using a common parser to extract the label from both the
'btrfs filesystem show' and 'btrfs-show' commands which can read the
label with and without enclosing single quotes.
Patch 2/4 - btrfs::read_label()
This patch changes the btrfs file system label parsing code to resolve
issue 1/2 by ignoring the exit status from the 'btrfs filesystem show'
command and relying on parsing the required information to determine
success or failure. Issue 2/2 is also resolved as described above.
Bug #733601 - Btrfs: Warnings and missing label with btrfs-progs 3.12
and 3.14
Issue 1/2 - GParted shows warnings for mounted btrfs file systems
'btrfs filesystem show /dev/PTN' command is used to query details of a
btrfs file system. When the file system is mounted the command reports
failed exit status 1, but only when using btrfs-progs v3.14 and v3.14.1.
This causes GParted to: (1) report warnings from the failed commands for
a mounted btrfs file system, (2) fail to determine file system usage
figures and (3) fail to display the mount point and busy indicator for
non-mounting devices in multi-device btrfs file systems. The label is
also read using the secondary blkid method via the FS_Info cache.
Failed exit status 1 when the btrfs file system is mounted:
# fgrep sdb1 /proc/mounts
/dev/sdb1 /mnt/1 btrfs rw,seclabel,relatime,space_cache 0 0
# btrfs filesystem show /dev/sdb1
Label: 'test1-btrfs' uuid: 033e6b07-ee6a-4620-a585-8580a2b83275
Total devices 1 FS bytes used 192.00KiB
devid 1 size 2.00GiB used 240.75MiB path /dev/sdb1
Btrfs v3.14.1
# echo $?
1
Successful exit status 0 when the btrfs file system is unmounted:
# umount /dev/sdb1
# btrfs filesystem show /dev/sdb1
Label: 'test1-btrfs' uuid: 033e6b07-ee6a-4620-a585-8580a2b83275
Total devices 1 FS bytes used 192.00KiB
devid 1 size 2.00GiB used 240.75MiB path /dev/sdb1
Btrfs v3.14.1
# echo $?
0
Fix by ignoring the exit status of the 'btrfs filesystem show' command
and rely on parsing the required information to determine success or
failure. The output from the older 'btrfs-show' command is almost
identical so the same code will parse it in all cases.
Patch 1/4 - btrfs::read_uuid()
This patch changes the secondary method used to read the btrfs UUID to
resolve issue 1/2 as described above.
Bug #733601 - Btrfs: Warnings and missing label with btrfs-progs 3.12
and 3.14
While one partition is busy, reformat another partition from the command
line. Afterwards parted/libparted still detects the original file
system and GParted shows errors from the file system specific tools
reporting the new file system doesn't exist. Only limitation is that
the new new file system must be recognised by libparted (or by GParted's
fallback file system signature detection).
Case #1, File system reformatting:
# parted /dev/sdb print
Model: ATA SAMSUNG SSD UM41 (scsi)
Disk /dev/sdb: 8012MB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
1 1049kB 2149MB 2147MB primary ext2
2 2149MB 4296MB 2147MB primary ext2
# mount | fgrep sdb
/dev/sdb1 on /mnt/1 type ext2 (rw)
# mkfs.xfs -f /dev/sdb2
# blkid /dev/sdb2
/dev/sdb2: UUID="c31823a2-b81b-46fa-8246-0a59695e4834" TYPE="xfs"
# parted /dev/sdb print
Model: ATA SAMSUNG SSD UM41 (scsi)
Disk /dev/sdb: 8012MB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
1 1049kB 2149MB 2147MB primary ext2
2 2149MB 4296MB 2147MB primary ext2
# e2label /dev/sdb2
e2label: Bad magic number in super-block while trying to open /dev/sdb2
Couldn't find valid filesystem superblock.
# dumpe2fs /dev/sdb2
dumpe2fs 1.41.12 (17-May-2010)
dumpe2fs: Bad magic number in super-block while trying to open /dev/sdb2
Couldn't find valid filesystem superblock.
Case #2, Removing device from multi-device btrfs:
# btrfs filesystem show /dev/sdb1
Label: none uuid: a05db434-efd5-4e8c-902f-05f89a88b610
Total devices 2 FS bytes used 156.00KB
devid 2 size 2.00GB used 512.00MB path /dev/sdb2
devid 1 size 2.00GB used 240.75MB path /dev/sdb1
# mount /dev/sdb1 /mnt/1
# btrfs device delete /dev/sdb2
# btrfs filesystem show /dev/sdb1
Label: none uuid: a05db434-efd5-4e8c-902f-05f89a88b610
Total devices 1 FS bytes used 92.00KB
devid 1 size 2.00GB used 714.25MB path /dev/sdb1
# btrfs filesystem show /dev/sdb2
and GParted reports this error for partition /dev/sdb2:
Unable to read the contents of this file system!
Because of this some operations may be unavailable.
The cause might be a missing software package.
The following list of software packages is required for btrfs
file system support: btrfs-tools.
This is another case of libparted reading from the whole disk device
(/dev/sdb) yet the file system tools use the partition specific block
device (/dev/sdb2), and the Linux buffer cache not providing cache
coherency. Previous scenario was fixed with:
797f0b8eeb
Flush device after wiping a file system (#688882)
This affects libparted 2.0 to 3.1 inclusive and is fixed by:
http://git.savannah.gnu.org/cgit/parted.git/commit/?id=fb99ba5ebd0dc34204fc9f1014131d5d494805bc
Revert "linux-commit: do not unnecessarily open partition device nodes"
Fix by calling ped_device_sync() to guarantee cache coherency for each
device during scanning.
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
With linux 3.5 and later, the device used to mount a btrfs file system
is updated in /proc/mounts when the previous mounting device is removed
from the file system. Most recent distributions make /etc/mtab a
symbolic link to /proc/mounts. However some still have /etc/mtab as a
plain file only updated by mount and umount, thus showing the old device
name which is no longer part of the file system.
On Ubuntu 13.10, which has /etc/mtab as a plain file managed by mount
and umount:
# mkfs.btrfs /dev/sdb1
# mount /dev/sdb1 /mnt/1
# btrfs device add /dev/sdb2 /mnt/1
# btrfs device delete /dev/sdb1 /mnt/1
# sync
# btrfs filesystem show /dev/sdb1
# btrfs filesystem show /dev/sdb2
Label: none uuid: e47775a6-e5ad-4fb4-9ea4-1570aa5b4009
Total devices 2 FS bytes used 28.00KB
devid 2 size 2.00GB used 272.00MB path /dev/sdb2
# fgrep btrfs /proc/mounts
/dev/sdb2 /mnt/1 btrfs rw,relatime,space_cache 0 0
# ls -l /etc/mtab
-rw-r--r-- 1 root root 842 Apr 15 19:41 /etc/mtab
# fgrep btrfs /etc/mtab
/dev/sdb1 /mnt/1 btrfs rw 0 0
This causes GParted to report /dev/sdb1 as busy and mounted at /mnt/1
when it is no longer mounted. This effects recent releases of Ubuntu,
13.04, 13.10 and 14.04.
Either /etc/mtab is a symlink and is identical to /proc/mounts or
/etc/mtab is a plain file with wrong information. Fix by not reading
mounted file systems from /etc/mtab.
However old distributions only contain 'rootfs' and '/dev/root' device
names for the / (root) file system with '/dev/root' being a block device
rather than a symlink to the true device. For example from CentOS 5.x:
# fgrep ' / ' /proc/mounts
rootfs / rootfs rw 0 0
/dev/root / ext3 rw,data=ordered 0 0
# ls -l /dev/root
brw------- 1 root root 8, 3 Jun 4 2013 /dev/root
This prevents identification, and therefore busy detection, of the
device containing the / (root) file system. Used to read /etc/mtab to
get the root file system device name.
# fgrep ' / ' /etc/mtab
/dev/sda3 / ext3 rw 0 0
# ls -l /dev/sda3
brw-r----- 1 root disk 8, 3 Jun 4 2013 /dev/sda3
As per commit:
409096f739
improved scanning for root mountpoint (/) ...
but, as discussed above, this contains an out of date device name after
the mounting device has been dynamically removed from a multi-device
btrfs, thus identifying the wrong device as busy. Instead fall back to
reading mounted file systems from the output of the mount command, but
only when required.
# mount | fgrep ' / '
/dev/sda3 on / type ext3 (rw)
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
In a number of cases there may be no btrfs device cache entry. Some of
the reasons why this can occur:
1) Mounting device removed from btrfs on linux <= 3.4 so old mount
point in /proc/mounts;
2) btrfs and btrfs-show commands don't exist;
3) btrfs or btrfs-show command returned non-zero exit status;
4) get_cache_entry() failed to parse output from btrfs filesystem show
or btrfs-show.
Without a valid btrfs device cache entry, busy detection for all member
devices fails. Search the GParted internal mounted partitions map as
the fallback busy detection method. This can only determine if the
mounting device is mounted or not, not any of the other members of a
multi-device btrfs file system.
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
GParted doesn't specify the devid when resizing a btrfs file system, so
the kernel defaults to resizing devid 1. On a multi-device btrfs this
may not be the same partition which GParted is resizing. This will
result in file system truncation and corruption. Shrinking the wrong
partition example:
1) Create a btrfs file system spanning 2 partitions:
# mkfs.btrfs /dev/sdb1 /dev/sdb2
# btrfs filesystem show /dev/sdb1
Label: none uuid: 41654265-9840-45c4-aca1-55989da358d6
Total devices 2 FS bytes used 112.00KiB
devid 1 size 2.00GiB used 437.50MiB path /dev/sdb1
devid 2 size 2.00GiB used 417.50MiB path /dev/sdb2
2) Resize /dev/sdb2 down to 1 GiB using GParted. This command was
run:
btrfs filesystem resize 1048576K /tmp/gparted-ddyGRh
which resized devid 1 (/dev/sdb1) to 1 GiB:
# btrfs filesystem show /dev/sdb1
Label: none uuid: 41654265-9840-45c4-aca1-55989da358d6
Total devices 2 FS bytes used 256.00KiB
devid 1 size 1.00GiB used 437.50MiB path /dev/sdb1
devid 2 size 2.00GiB used 417.50MiB path /dev/sdb2
but GParted instead resized /dev/sdb2 to 1 GiB:
# sfdisk -s /dev/sdb1
2097152
# sfdisk -s /dev/sdb2
1048576
Even on a single device btrfs devid 1 may no longer exist if the file
system has had the initial device removed from it. Example:
1) Create a single btrfs file system, add a second device and
remove the first:
# mkfs.btrfs /dev/sdb1
# mount /dev/sdb1 /mnt/1
# btrfs device add /dev/sdb2 /mnt/1
# btrfs device remove /dev/sdb1 /mnt/1
# umount /mnt/1
# btrfs filesystem show /dev/sdb2
Label: none uuid: 2cbf3ac3-1344-472a-a0c7-1476d23bdc9f
Total devices 1 FS bytes used 256.00KiB
devid 2 size 2.00GiB used 480.00MiB path /dev/sdb2
2) Again resize /dev/sdb2 down to 1 GiB using GParted. This
command was run:
btrfs filesystem resize 1048576K /tmp/gparted-ddyGRh
but it failed with:
ERROR: unable to resize 'tmp/gparted-lEyGaY' - No such device
A more informative error message was written to syslog:
# tail -1 /var/log/messages
Mar 12 14:15:01 localhost kernel: btrfs: resizer unable to find device 1
This is with Linux kernel 3.13.5 on Fedora 20, circa March 2014.
Fix by specifying the devid when resizing (part of) a btrfs file system.
Example command specifying devid 2:
btrfs filesystem resize 2:1048576K /tmp/1
This will always work because it is the kernel which interprets the
devid colon size parameter and has always done so since btrfs was first
added to the kernel in version 2.6.32 [1].
Reference:
[1] linux v2.6.32 fs/btrfs/ioctl.c btrfs_ioctl_resize()
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/fs/btrfs/ioctl.c?id=v2.6.32#n578
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
Currently GParted fails to report the usage of a multi-device btrfs file
system if it is mounted or if the used space is larger than the size of
an individual member device. When GParted does display usage figures it
also incorrectly reports the file system wide used figure against every
member device.
Mounted case:
statvfs() provides an FS size which is larger than any individual
device so is rejected. See:
GParted_Core::mounted_set_used_sectors()
Utils::get_mounted_filesystem_usage()
partition .set_sector_usage()
Unmounted case, FS used > device size:
FS used figure is larger than any individual device so free space is
calculated as a negative number and rejected. See:
btrfs::set_used_sectors()
Btrfs has a volume manager layer within the file system which allows it
to provide multiple levels of data redundancy, RAID levels, and use
multiple devices both of which can be changed while the file system is
mounted. To achieve this btrfs has to allocate space at two different
level: (1) chunks of 256 MiB or more at the volume manager level; and
(2) extents at the file data level.
References:
* Btrfs: Working with multiple devices
https://lwn.net/Articles/577961/
* Btrfs wiki: Glossary
https://btrfs.wiki.kernel.org/index.php/Glossary
This makes the question of how much disk space is being used in an
individual device a complicated question to answer. Further, the
current btrfs tools don't provide the required information.
Btrfs filesystem show only provides space usage information at the chunk
level per device. At the file extent level only a single figure for the
whole file system is provided. It also reports size of the data and
metadata being stored, not the larger figure of the amount of space
taken after redundancy is applied. So it is impossible to answer the
question of how much disk space is being used in an individual device.
Example output:
Label: none uuid: 36eb51a2-2927-4c92-820f-b2f0b5cdae50
Total devices 2 FS bytes used 156.00KB
devid 2 size 2.00GB used 512.00MB path /dev/sdb2
devid 1 size 2.00GB used 240.75MB path /dev/sdb1
Fix by guesstimating the per device used figure as the fraction of the
file system wide extent usage based on chunk usage per device.
Calculation:
ptn fs used = total fs used * devid used / sum devid used
Positives:
1) Per device used figure will correctly be between zero and allocated
chunk size.
Known inaccuracies:
[for single and multi-device btrfs file systems]
1) Btrfs filesystem show reports file system wide file extent usage
without considering redundancy applied to that data. (By default
btrfs stores two copies of metadata and one copy of data).
2) At minimum size when all data has been consolidated there will be a
few partly filled chunks of 256 MiB or more for data and metadata of
each storage profile (RAID level).
[for multi-device btrfs file systems only]
3) Data may be far from evenly distributed between the chunks on
multiple devices.
4) Extents can be and are relocated to other devices within the file
system when shrinking a device.
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
Linux can only show a single device name in /proc/mounts and /etc/mtab
for each mounted btrfs, even if it is a multi-device file system. So
GParted only shows a mount point for one of the devices in the btrfs, no
matter how many devices are part of the file system.
# mkfs.btrfs /dev/sdb1 /dev/sdb2
# btrfs filesystem show /dev/sdb1
Label: none uuid: 36eb51a2-2927-4c92-820f-b2f0b5cdae50
Total devices 2 FS bytes used 156.00KB
devid 2 size 2.00GB used 512.00MB path /dev/sdb2
devid 1 size 2.00GB used 240.75MB path /dev/sdb1
# mount /dev/sdb1 /mnt/1
# grep btrfs /proc/mounts
/dev/sdb1 /mnt/1 btrfs rw,seclabel,relatime,ssd,space_cache 0 0
GParted only shows the mount point for /dev/sdb1 as /mnt/1, but nothing
for /dev/sdb2.
Make GParted report the same mount point for all devices included in a
multi-device btrfs file system.
Add btrfs specific get_mount_device() method to report the mounting
device, if any, for the btrfs file system in the occupying the device in
question. Uses the existing cache of btrfs file system device
membership. Also extract common code from GParted_Core::
set_mountpoints() into set_mountpoints_helper().
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
Busy detection of file systems works by checking if the device is
mounted (appears in the mount_info map). For a multi-device btrfs file
system this will only report one of the devices as busy, not all of
them.
# btrfs filesystem show /dev/sdb1
Label: none uuid: 36eb51a2-2927-4c92-820f-b2f0b5cdae50
Total devices 2 FS bytes used 156.00KB
devid 2 size 2.00GB used 512.00MB path /dev/sdb2
devid 1 size 2.00GB used 240.75MB path /dev/sdb1
# mount /dev/sdb1 /mnt/1
# grep btrfs /proc/mounts
/dev/sdb1 /mnt/1 btrfs rw,seclabel,relatime,ssd,space_cache 0 0
GParted will only report /dev/sdb1 as busy, but not /dev/sdb2.
Add btrfs specific is_busy() method which reports the device as busy if
any of the devices in the btrfs file system are mounted. This uses a
cache which maps device membership in all btrfs file systems. The cache
is cleared on GParted refresh and incrementally populated as each btrfs
partition is checked for busy status.
WARNING:
Removal of the mounting device from a btrfs file system makes it
impossible to determine whether the file system is mounted or not for
linux <= 3.4. This is because /proc/mounts continues to show the old
device which is no longer a member of the file system.
# btrfs device delete /dev/sdb1 /mnt/1
# sync
# grep btrfs /proc/mounts
/dev/sdb1 /mnt/1 btrfs rw,seclabel,relatime,ssd,space_cache 0 0
# btrfs filesystem show /dev/sdb1
# btrfs filesystem show /dev/sdb2
Label: none uuid: 36eb51a2-2927-4c92-820f-b2f0b5cdae50
Total devices 1 FS bytes used 28.00KB
devid 2 size 2.00GB used 1.02GB path /dev/sdb2
Fixed in linux 3.5 by commit:
Btrfs: implement ->show_devname
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=9c5085c147989d48dfe74194b48affc23f376650
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
Add static member function GParted_Core::is_dev_mounted() so that other
modules can determine if a particular partition contains a mounted file
system or not.
Make it a static member function so that it can be called without
needing the gparted_core object. Extend to make the group of
manipulated variables (mount_info, fstab_info) and manipulating
functions (init_maps(), read_mountpoints_from_file(),
read_mountpoints_from_file_swaps(), get_all_mountpoints()) static too.
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
Previous commit:
Make partition busy detection method selectable per file system
(#723842)
causes GParted to no longer detect an unknown mounted file system as
busy.
This is needed because there are many less popular file systems which
Linux can mount but GParted doesn't recognise, such as: Acorn ADFS,
AFFS, BeFS, BFS, etc. Encountering an unrecognised mounted device can
also occur with btrfs on Linux 3.4 and earlier when the mounting device
is deleted from a multi-device btrfs file system. Happens because
/proc/mounts continues to show the original mounting device even after
that device was deleted from the file system.
# mkfs.btrfs /dev/sdb1
# mount /dev/sdb1 /mnt/1
# btrfs device add /dev/sdb2 /mnt/1
# btrfs device delete /dev/sdb1 /mnt/1
# btrfs filesystem sync /mnt/1
# grep btrfs /proc/mounts
/dev/sdb1 /mnt/1 btrfs rw,seclabel,relatime,ssd,space_cache 0 0
# blkid /dev/sdb1
# btrfs filesystem show /dev/sdb1
# blkid /dev/sdb2
/dev/sdb2: UUID="9c75647c-217a-4718-bcc7-f3ccd8cc7dc6" UUID_SUB="b5d43630-80d4-42ac-b406-185e97cd5bbe" TYPE="btrfs"
# btrfs filesystem show /dev/sdb2
Label: none uuid: 9c75647c-217a-4718-bcc7-f3ccd8cc7dc6
Total devices 1 FS bytes used 28.00KB
devid 2 size 2.00GB used 1.02GB path /dev/sdb2
Reinstate the search for mounted partitions for busy detection even for
unrecognised file system types.
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
GParted's primary inbuilt busy detection method is "is the partition
mounted?". A custom method is used for LVM2 PV because its not a
mounted file system.
Make busy detection selectable per file system type.
.fs.busy = FS::NONE (default)
No busy detection.
.fs.busy = FS::GPARTED
Use internal GParted method which checks if the partition is
mounted.
.fs.busy = FS:EXTERNAL
Call the file system type's member function is_busy().
LVM2 PV busy detection changes from a special case to just electing to
call the lvm2_pv::is_busy() method. Linux Software RAID remains a
special case because it's only recognised, but not otherwise supported.
Bug #723842 - GParted resizes the wrong filesystem (does not pass the
devid to btrfs filesystem resize)
Fragment of debugging and valgrind output:
D: tid=2193 main()
...
D: tid=2202 GParted_Core::set_devices_thread()
...
D: tid=2202 Utils::execute_command(command="dumpe2fs -h /dev/sda1", output, error, use_C_locale=1)
D: tid=2202 this=0x13fef4a0 PipeCapture::PipeCapture()
D: tid=2202 this=0x13fef4f0 PipeCapture::PipeCapture()
D: tid=2202 this=0x13fef4a0 PipeCapture::connect_signal()
D: sourceid=77
D: tid=2202 this=0x13fef4f0 PipeCapture::connect_signal()
D: sourceid=78
D: tid=2193 data=0x13fef4a0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4a0 PipeCapture::OnReadable()
D: signal_update.emit()
D: return true
D: tid=2193 data=0x13fef4f0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4f0 PipeCapture::OnReadable()
D: signal_update.emit()
D: return true
D: tid=2193 data=0x13fef4a0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4a0 PipeCapture::OnReadable()
D: signal_update.emit()
D: return true
D: tid=2193 data=0x13fef4f0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4f0 PipeCapture::OnReadable()
D: signal_eof.emit()
D: return false
D: (!rc) &(pc->sourceid)=0x13fef518
D: tid=2193 data=0x13fef4a0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4a0 PipeCapture::OnReadable()
D: signal_update.emit()
D: return true
D: tid=2193 data=0x13fef4a0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4a0 PipeCapture::OnReadable()
D: signal_update.emit()
D: return true
D: tid=2193 data=0x13fef4a0 PipeCapture::_OnReadable()
D: tid=2193 this=0x13fef4a0 PipeCapture::OnReadable()
D: signal_eof.emit()
D: tid=2202 this=0x13fef4f0 PipeCapture::~PipeCapture()
D: sourceid=0
D: tid=2202 this=0x13fef4a0 PipeCapture::~PipeCapture()
D: sourceid=77
D: return false
D: (!rc) &(pc->sourceid)=0x13fef4c8
==2193== Thread 1:
==2193== Invalid write of size 4
==2193== at 0x490580: GParted::PipeCapture::_OnReadable(_GIOChannel*, GIOCondition, void*) (PipeCapture.cc:56)
==2193== by 0x38662492A5: g_main_context_dispatch (gmain.c:3066)
==2193== by 0x3866249627: g_main_context_iterate.isra.24 (gmain.c:3713)
==2193== by 0x3866249A39: g_main_loop_run (gmain.c:3907)
==2193== by 0x3D7FD45C26: gtk_main (gtkmain.c:1257)
==2193== by 0x469743: GParted::GParted_Core::set_devices(std::vector<GParted::Device, std::allocator<GParted::Device> >&) (GParted_Core.cc:155)
==2193== by 0x4A78F1: GParted::Win_GParted::menu_gparted_refresh_devices() (Win_GParted.cc:1259)
==2193== by 0x4A7886: GParted::Win_GParted::on_show() (Win_GParted.cc:1253)
==2193== by 0x3D82B2009C: Gtk::Widget_Class::show_callback(_GtkWidget*) (widget.cc:3855)
==2193== by 0x3867210297: g_closure_invoke (gclosure.c:777)
==2193== by 0x3867221B86: signal_emit_unlocked_R (gsignal.c:3516)
==2193== by 0x386722A0F1: g_signal_emit_valist (gsignal.c:3330)
==2193== Address 0x13fef4c8 is not stack'd, malloc'd or (recently) free'd
==2193==
PipeCapture.cc (with debugging):
46 gboolean PipeCapture::_OnReadable( GIOChannel *source,
47 GIOCondition condition,
48 gpointer data )
49 {
50 std::cout << "D: tid=" << (long int)syscall(SYS_gettid) << " data=" << data << " PipeCapture::_OnReadable()" << std::endl;
51 PipeCapture *pc = static_cast<PipeCapture *>(data);
52 gboolean rc = pc->OnReadable( Glib::IOCondition(condition) );
53 if (!rc)
54 {
55 std::cout << "D: (!rc) &(pc->sourceid)=" << &(pc->sourceid) << std::endl;
56 pc->sourceid = 0;
57 }
58 return rc;
59 }
The use after free across threads only happens when an external program
is being executed from a thread other than the main() thread. This is
because by default glib registered callbacks are run by the glib main
loop, which is only called from the main() thread with Gtk::Main::run().
Event sequence:
tid=2193 tid=2202
main()
...
GParted_Core::set_devices()
Glib::Thread::create(... set_devices_thread ...)
Gtk::Main::run() GParted_Core::set_devices_thread()
...
Utils::execute_command("dumpe2fs ... /dev/sda1" ...)
Glib::spawn_async_with_pipes()
PipeCapture outputcapture(out, output)
outputcapture.connect_signal()
//Glib main loop runs callback
PipeCapture::_OnReadable()
pc->OnReadable()
//output read
signal_update.emit()
return true
...
//Glib main loop runs callback
PipeCapture::_OnReadable()
pc->OnReadable()
//eof reached
[1] signal_eof.emit()
return status.exit_status
[2] PipeCapture::~PipeCapture()
[3] return false
[4] pc->sourceid = 0
What is happening is that the PipeCapture destructor [2] is running in
the set_devices_thread() thread and freeing the object's memory as soon
as signal_eof.emit() [1] has been called. Then signal_eof.emit()
returns back to OnReadable() which then returns false [3] back to the
_OnReadable() callback function which then assigns 0 to sourceid member
variable [4] in the already freed object, detected by valgrind as:
Invalid write of size 4
at ... GParted::PipeCapture::_OnReadable(...) (PipeCapture.cc:56)
This is happening because PipeCapture member variable sourceid is being
saved, in a different thread, just so the _OnReadable() callback can be
removed. However a glib IOChannel callback, type GIOFunc(), returning
false will be automatically removed.
GLib Reference Manual 2.26 / IO Channels
https://developer.gnome.org/glib/2.26/glib-IO-Channels.html#GIOFunc
GIOFunc()
Returns : the function should return FALSE if the event source
should be removed
Therefore fix by just not saving the event sourceid at all, and not
calling g_source_remove() to manually remove the callback, but instead
letting glib automatically remove the callback when it returns false.
Bug #731752 - Write after free cross thread race in
PipeCapture::_OnReadable()
OperationDetail was storing its children in a std::vector. This means they
can be moved around in memory arbitrarily, going through indeterminate
lifetimes. This is generally a bad thing for any non trivial object and
in the case of OperationDetail, it created havoc with the way it maintains
pointers between parent/child objects for signal connections. It will now
keep only pointers to children in a std::vector instead, so their lifetime
can be controlled, fixing various crashes.
Bug 729139 - Refactor OperationDetail to address random behavior
Returning false from the OnReadable callback causes the source to
be destroyed, but PipeCapture::~PipeCapture was destroying it a
second time. Prevent this by zeroing out the sourceid.
Bug 729800 - Prevent GSource double-destroy warning messages
Place the LVM2 member names in a single string, separated by new lines,
so that they can be selected all together by the user. This was not
previously possible when each member name was placed in a separate
widget.
A LVM volume group can contain more than one member and comprise a
multi-line entry. As such set the "Members" field header and value
label pair to top vertical alignment.
As part of the work on bug 652044 - uses deprecated APIs, selectable
vertical alignment was defaulted to ALIGN_CENTER for all labels. The
relevant commits can be viewed in comment 26 of said bug report.
https://bugzilla.gnome.org/show_bug.cgi?id=652044#c26
For multi-line labels a vertical ALIGN_CENTER value is not
consistently aesthetically pleasing. This becomes obvious when a
single-line heading label is paired with a multi-line value label.
To improve the aesthetics, a vertical alignment of ALIGN_TOP is
preferred.
Hence re-add the ability to optionally specify a vertical alignment for
labels. If a yalign value is not specified a default vertical alignment
of ALIGN_CENTER is used.
With the volume group name displayed immediately below the status field,
the vgname is no longer needed in the status field.
Bug 690542 - Partition Information Dialog Warning not readable
When the partition information dialog is resized, the data below the
partition graphic would remain left aligned. Improve the aesthetics by
horizontally centering the data in the window similar to the partition
graphic.
Bug 690542 - Partition Information Dialog Warning not readable
Placing the LVM2 member names in a string list enables the member names
to be selected all at once by a user. This was not possible with
separate label widgets for each LVM2 member name.
Bug 690542 - Partition Information Dialog Warning not readable
Make the dialog resizable, add a vertical scrollbar to the information
and messages section, and set the initial height to ensure the dialog
fits entirely on an 800x600 screen.
A default height is required because some window managers, such as
fluxbox used in GParted Live, only permit resizing the height by using
the bottom corners of the dialog. If the dialog is too large for the
screen then the user would not be able to resize it.
Note that two default initial heights are used in an effort to minimize
the amount of extra whitespace.
Bug 690542 - Partition Information Dialog Warning not readable
On some desktop environments, such as Unity on Ubuntu 13.04, the
partition information dialog can display the used/unused/unallocated
field values overlapping with the "(##%)" percentage value. The problem
is pronounced for values with 4 digits to the left of the decimal (for
example a 1024 MiB partition ). To address this problem, these
percentages are placed in their own column.
Bug 690542 - Partition Information Dialog Warning not readable
Organize the partition information field & value areas into two columns
to minimize the amount of vertical space required.
This is part of a series of changes to enable viewing all partition
information details on a minimum 800x600 display.
Part of Bug 690542 - Partition Information Dialog Warning not readable
This is part of a series of changes to enable viewing all partition
information details on a minimum 800x600 display.
Part of Bug 690542 - Partition Information Dialog Warning not readable
This is part of a series of changes to enable viewing all partition
information details on a minimum 800x600 display.
Part of Bug 690542 - Partition Information Dialog Warning not readable
The code used to unnecessarily destroy and re-create the file system
objects on every scan for file system support tools.
Instead only create the file system objects once and just call each
object's get_filesystem_support() method on each rescan.
Prior to commit:
1f3b11748e
Remove GParted_Core::p_filesystem (#683149)
set_proper_filesystems() used to set GParted_Core::p_filesystem member
variable to one of the FileSystem objects, but that was just treating it
like a local variable. After the commit local variables named
p_filesystem were used where required and set_proper_filesystem() became
a function which did nothing other than call get_filesystem_object().
Now remove set_proper_filesystem() altogether and use
get_filesystem_object() in its place.
Comment is outdated after GParted stopped using ped_partition_is_busy()
for busy partition detection of normal and logical partitions since
commit:
4202992063
Fix false busy detection of unusual case with Linux Software RAID (#712533)
The String::ucompose method is for easy, i18n-friendly composition of
strings with Gtkmm. From past experience, String::ucompose should not
be used to build command lines due to differences in how locales handle
number translation. More specifically, not all locales use number
separators (spaces, commas or periods) and the decimal separator
(period or comma) in the same way.
The problem with using String::ucompose for command lines was
originally encountered and corrected in the following commit:
Fix attempt data rescue fail to open read-only view (#673246)
e494eca1f7
Use e2image features added in e2fsprogs 1.42.9 to move/copy
an ext[234] file system more efficiently by skipping unused blocks.
Fall back to the internal copy algorithm if e2image is not found
or does not support move/copy.
Bug #721516 - Use e2image to move/copy ext[234] filesystems
Work on bug 701075 inadvertently introduced a problem where ext2/3/4,
resierfs, and ntfs file systems were rounding the file system size up
to the nearest kiB.
The problem was discovered when a user resized a partition by moving
only the start boundary to the right thereby shrinking the partition.
In the situation where the resulting partition size was not an integer
kiB value (such as might occur on a 512 byte per sector device) the
file system size was rounded up to the nearest kiB. This resulted in
a file system size being set one sector larger than the partition
size.
This problem was introduced with the following commit:
Shrink file systems to exact size (#701075)
3461413d28
The fix chosen for this problem involved removing the rounding logic.
Bug 723543 - Shrink ext2/3/4 results in attempt to set partition
smaller than file system
Only recognises partitions containing BitLocker Disk Encryption content.
No other actions are supported.
Bug #723232 - BitLocker Disk Encryption not recognised
Restore the order of the source files so that they are once again
compiled in order A-Z, a-z. Order is obtained with:
fgrep .cc src/Makefile.am | LANG=C sort
fgrep .h include/Makefile.am | LANG=C sort
Make the dialog resizable, add a vertical scrollbar and set the minimum
(and therefore initial) height to 500 pixels. This is so that the
dialog entirely fits on an 800x600 screen, thus allowing the rescan
button to be pressed.
100 pixel difference is to account for the size of the top and bottom
GNOME 2 panels and two sets of title bars. Two sets of title bars
because the window manager tries to place the top of dialog title bars
in line with the bottom of the main window title bar.
Bug #342682 - too much information in 'features' dialog
Use padding to indent the legend narrative to better show hierarchy and
association with the "Legend" category as recommended in the:
GNOME HIG 2.2.3 / Visual Design / Window Layout
https://developer.gnome.org/hig-book/stable/design-window.html.en
Padding the legend narrative widget also deliberately increases the
spacing to the legend icons too.
Bug #342682 - too much information in 'features' dialog
Make the legend always shown, ready for when the dialog is resizable.
Change the widget containing the legend from an expander to a frame
widget. Set the frame to be borderless using a bold label as
recommended in the GNOME Human Interface Guidelines 2.2.1 / Controls /
Frames and Separators.
https://developer.gnome.org/hig-book/2.32/controls-frames.html.en
Bug #342682 - too much information in 'features' dialog
The DialogFeatures::DialogFeatures() constructor used four levels of
nested code blocks and reused the same variable names at multiple
levels. The variable hbox was used at three of the four levels. This
was done by commit:
91b971691d
Bug #634090: Change some attributes to local variables
In my opinion this made the code hard to understand. Give every local
variable a unique and meaningful name and remove the nested code blocks.
This is part of parent bug:
Bug #721455 - Obsolete info in license text on multiple modules
and GNOME Goal:
https://wiki.gnome.org/Initiatives/GnomeGoals/Proposals
* verify all source files to make sure they have a license and a
copyright, and that both are up-to-date
Bug #721565 - License text contains obsolete FSF postal address
Write a loop partition table to a disk and then try to create a new
partition on that disk crashes GParted.
Analysis:
1) Loop partition table, actually just the signature / label
"GNU Parted Loopback 0", is written to the disk.
2) create_partition() calls ped_disk_app_partition() to create the
partition which, arguably erroneously, clears the loop signature.
3) In erase_filesystem_signatures(), get_device_and_disk() fails because
there's no longer a loop signature for libparted to recognise. This
leaves lp_device, lp_disk and lp_partition = NULL, but incorrectly
overall_success = true.
4) Overall_success = true allows the signature erasure loop to run and
evaluate lp_device->sector_size, dereferencing a NULL pointer and
crashing GParted.
Fix erase_filesystem_signatures() to correctly handle failure from
get_device_and_disk(), broken by this commit:
246e05559d
Clear btrfs file system super block mirror copies too (#705426)
Bug #721988 - Crash when creating new partition on disk with loop label
In the Create Partition Table dialog display the entries in the combobox
in order.
Previously the default of MSDOS or GPT was moved to the first item in
the combobox. Now the partition table types remain in order with just
either MSDOS or GPT being selected as as the default as required.
The partition table types are displayed in the order supplied by
libparted, which is alphabetic except with "loop" last.
Bug #711098 - Default partition table can not handle > 2 TiB disks
MSDOS partition table is limited to addressing 2^32 sectors, limiting
disks using 512 byte sectors to 2 TiB in size. Fdisk reports the
following warning on disks 2 TiB and larger.
# truncate -s 2T /var/tmp/loop-2T
# losetup /dev/loop0 /var/tmp/loop-2T
# fdisk /dev/loop0
WARNING: The size of this disk is 2.2 TB (2199023255552 bytes).
DOS partition table format can not be used on drives for volumes
larger than (2199023255040 bytes) for 512-byte sectors. Use parted(1) and GUID
partition table format (GPT).
(Fdisk arguably reports this warning one sector too early. Anyway for
safety and consistency GParted will use this limit too). Continue to
use MSDOS as the default partition table type for disks smaller than 2
TiB and use GPT as the default for disks 2 TiB and larger. This
maximises compatibility.
Also remove the advanced expander and always show the partition table
list box.
Bug #711098 - Default partition table can not handle > 2 TiB disks
When the kernel version as stated in /proc/version did not have at least
three numbers separated by periods, the version would fail to be read.
Sample /proc/version to demonstrate problem:
Linux version 3.10-3-686-pae ...
The Linux kernel will always have a major number and a minor number
separated by a period. There is likely also a patch version number too
that would be preceded by a period. This enhancement will read 2 and 3
component Linux kernel versions.
GParted doesn't report extended partitions as busy if it contains only
busy LVM2 PV or SWRAID logical partitions. Libparted's
ped_partition_is_busy() only detects mounted file systems and swap space
as busy, not active LVM2 PVs or SWRAID members. This is as of libparted
3.1 and earlier.
Fix by determining the busy status of an extended partition based solely
on the busy status of the logical partitions it contains. This makes it
unnecessary to check for mounted DMRAID logical partitions or call
ped_partition_is_busy() for extended partitions.
Bug #712533 - Partitions not detected as busy inside Software RAID on
some distros
In this unusual setup of having one Linux Software RAID array
partitioned and a second array not partitioned, /dev/md1p1 partition in
the first array is falsely reported as busy when the file system filling
the second array /dev/md2 is mounted.
Partition RAID File System Mount mount
--------- ----------------- ---------------- -----------
/dev/sdb1 SWRAID /dev/md1
/dev/md1p1 ext4,Label=test1 (Unmounted)
/dev/sdb2 SWRAID /dev/md2 ext4,Label=test2 /mnt/2
This again affects Debian 6, Debian 7 and RHEL/CentOS 6.x and is as a
result of the this bug fixed in libparted 2.4:
http://git.savannah.gnu.org/cgit/parted.git/commit/?id=db52898c214310dab4ed84e6cb2f9ffc8c3aa502
linux: also detect "in-use" dmraid and scsi-Nth (N>=16) partition
Fix by not using libparted's ped_partition_is_busy() to determine if a
normal or logical partition is busy.
Bug #712533 - Partitions not detected as busy inside Software RAID on
some distros
On some distributions GParted doesn't detect when a file system is busy
and mounted when the partition is within a Linux Software RAID array.
Affected distributions include: Debian 6, Debian 7 and RHEL/CentOS 6.x.
This is because of a bug in libparted 2.3 and earlier. It was fixed by
this change in libparted 2.4:
http://git.savannah.gnu.org/cgit/parted.git/commit/?id=db52898c214310dab4ed84e6cb2f9ffc8c3aa502
linux: also detect "in-use" dmraid and scsi-Nth (N>=16) partition
Ubuntu 13.10 with libparted 2.3 is not affected because it contains a
distribution specific patch to correct it.
Fix by using GParted's internal mounted partitions map, mount_info, to
determine if a partition is busy, in addition to using libparted's
ped_partition_is_busy().
Bug #712533 - Partitions not detected as busy inside Software RAID on
some distros
Many file systems are capable of growing while mounted, and a few can
even shrink. This support must be explicitly enabled at configure time
with the --enable-online-resize flag and depends on a patched libparted.
Also requires kernel >= 3.6 for partition resizing, even if the
partition is in use (BLKPG_RESIZE_PARTITION).
Thanks to Mike Fleetwood for double check mark idea instead of a second
column to show the online grow/shrink.
Bug #694622 - Add support for online resize
In the Create New Partition dialog use ext4 as the default choice for
new file systems. It has been the preferred file system of
distributions for many years. Use ext3 and ext2 as second and third
choice defaults. This handles RHEL/CentOS 5.x which doesn't support
ext4.
Bug #711114 - Change default fs to ext4
Allow all partition warning messages to be selectable and copied. Merge
all the messages into a single Gtk::Label widget so that they can be
selected together.
Use a blank line between individual messages so that each message can be
distinguished. Therefore each message should be formatted as one or more
non-blank lines, with an optional trailing new line. This is true of
GParted internal messages and probably all external messages and errors
from libparted and executed commands too.
Bug #705596 - Partition Information Dialog - let user copy warnings
Read the contents of /proc/mdstat file to determine if a device is a
member of of an active RAID array.
$ cat /proc/mdstat
Personalities : [raid1]
md1 : active raid1 sda1[2] sdb1[3]
524224 blocks super 1.0 [2/2] [UU]
md2 : active raid1 sdb2[2] sda2[3](F)
5238720 blocks super 1.1 [2/1] [U_]
md3 : active raid1 sdb3[1]
10477440 blocks super 1.1 [2/1] [_U]
bitmap: 1/1 pages [4KB], 65536KB chunk
md4 : inactive sda4[0](S)
1048564 blocks super 1.2
unused devices: <none>
There are 5 example Linux Software RAID arrays, md1 to md5. All are
RAID1 mirrors with 2 members, in various states.
Array Members Status
md1 sda1, sdb2 Fully operational.
md2 sda2, sdb2 Member sda2 marked as faulty. (Device sda2 is
still in use).
md3 sda3, sdb3 Member sda3 has been removed. (Device sda3 is
not in use).
md4 sda4, sdb4 Incremental start of member sda4 only. (Neither
member devices is in use).
md5 sda5, sdb5 Array stopped. (Neither member device is in
use).
Also disable "Unmount" in the partition menu for active RAID array
members.
Bug #709640 - Linux Swap Suspend and Software RAID partitions not
recognised
Recognise in kernel, Linux Swap Suspend partitions. (When hibernated
the kernel write the RAM out to swap space and changes the magic string
from swap space to suspend). Recognition required either
libparted >= 1.8.8.1 or blkid from util-linux >= 2.15 or before that
blkid from e2fsprogs >= 1.39.
Recognise Linux Software RAID partitions. Recognition requires blkid
from util-linux >= 2.15.
Example:
# blkid /dev/sda10 /dev/sda11
/dev/sda10: ... TYPE="swsuspend"
/dev/sda11: ... TYPE="linux_raid_member"
# parted /dev/sda print
Model: ATA SAMSUNG HM500JI (scsi)
Disk /dev/sda: 500GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
...
10 361GB 362GB 1074MB logical swsusp
11 362GB 363GB 1074MB logical raid
Bug #709640 - Linux Swap Suspend and Software RAID partitions not
recognised
Step 3 of 3:
Now that all label widgets are created with ALIGN_LEFT and ALIGN_CENTER
alignment remove the x_align and y_align parameters from mk_label() and
always use this alignment. Also specify this alignment via floats
rather than enumerators, one of which was deprecated.
Bug #652044 - uses deprecated APIs
Step 2 of 3:
ALIGN_RIGHT was only used for the size figures in the GParted > Devices
device selector with radio buttons. Change to ALIGN_LEFT and adjust the
widget packing to maintain appearance.
Bug #652044 - uses deprecated APIs
Gtkmm release 2.24 in February 2011 deprecated the use of these
alignment enumerators superseding them with:
Deprecated enumerator Use instead
Gtk::ALIGN_LEFT Gtk::ALIGN_START
Gtk::ALIGN_RIGHT Gtk::ALIGN_END
Gtk::ALIGN_TOP Gtk::ALIGN_START
Gtk::ALIGN_BOTTOM Gtk::ALIGN_END
gtkmm: gtkmm Enums and Flags [2.24]
https://developer.gnome.org/gtkmm/2.24/group__gtkmmEnums.html
The deprecated alignment enumerators were only used in the creation of
Gtk::Label widgets by method Utils::mk_label(). In all but three cases
the alignment was left, center. So make all labels with alignment left,
center and then stop using the deprecated enumerators.
Step 1 of 3:
ALIGN_TOP was only used for these 2 labels:
"Path:" text in View > Device Information;
"Path:" text in Partition > Information.
Use of ALIGN_TOP makes no difference to these labels so just change to
ALIGN_CENTER.
Bug #652044 - uses deprecated APIs
Turn on the text progress bar for the resize2fs command. Looks like:
# resize2fs -p /dev/sda13 262144
resize2fs 1.41.12 (17-May-2010)
Resizing the filesystem on /dev/sda13 to 262144 (4k) blocks.
Begin pass 2 (max = 97177)
Relocating blocks XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
Begin pass 3 (max = 80)
Scanning inode table XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
Begin pass 4 (max = 5597)
Updating inode references XXXXXXXXX-------------------------------
Bug #709276 - Percentage indicator for subcommand
Add concept of cursor position within the current line, separate from
the end of the buffer. This is so that programs which output a text
progress bar using backspace, such as resize2fs -p, are displayed
correctly.
Bug #709276 - Percentage indicator for subcommand
All the code was there in btrfs::read_uuid() to read the UUID from the
btrfs-show command. It just needed enabling.
This is only relevant when: (1) the blkid command is unavailable as
GParted primarily reads the UUID through the FS_Info cache using the
blkid command, and (2) the btrfs command is unavailable and the btrfs
module uses the older btrfs-show command instead.
For active swap space read the usage from /proc/swaps. (Linux kernel
uses units of 1 KiB). By definition inactive swap space is 100% free.
$ cat /proc/swaps
Filename Type Size Used Priority
/dev/sda2 partition 5242876 430552 -1
Always set fs.read = FS::EXTERNAL even if /proc/swaps doesn't exist so
that an attempt is made to open the file generating a specific error, in
addition to the generic error.
open("/proc/swaps", O_RDONLY): No such file or directory
Unable to read the contents of this file system!
Because of this some operations may be unavailable.
The cause might be a missing software package.
The following list of software packages is required for linux-
swap file system support: util-linux.
Closes Bug #708107 - Usage of swap space is not reported
Enable the text progress bar and percentage complete from the e2fsck
command. Looks like:
# e2fsck -f -y -v -C 0 /dev/sda8
e2fsck 1.41.12 (17-May-2010)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
test8: |========================================= / 73.6%
Note that e2fsck deliberately brackets the progress bar with control-A
and control-B characters when it detects that it is not being run from a
terminal so that logsave(8) can exclude the progress bar from being
written to a log file. As GParted doesn't run e2fsck via a pty it
receives these control characters. Ignore them.
References:
Change e2fsck to bracket its progress bar output with ctrl-A and ctrl-B
http://git.kernel.org/cgit/fs/ext2/e2fsprogs.git/commit/?id=bc34d6be65cb93a65451ad209cfea2f98b03fd22
Don't print ^A and ^B chars when e2fsck is talking directly to a tty
http://git.kernel.org/cgit/fs/ext2/e2fsprogs.git/commit/?id=54a31a3b2e432c124aa03442f2983c4f4d4a974c
Closes Bug #697662 - Do not hide the progress of the tools used
Remove the -P, no progress bar option, from the ntfsresize command.
This allows the command to display the %age complete on its output which
is displayed in the operation details dialog, updated in real time.
Possible since:
Bug #685740 - Refactor to use asynchronous command execution
Bug #697662 - Do not hide the progress of the tools used
Currently the btrfs command outputs figures to 2 decimal places followed
by an SI multiplier, e.g. 1.00GB.
This patch to btrfs-progs has been included in the integration
repository and will likely be included in the official btrfs-progs
repository at some point. It changes btrfs-progs to use IEC
multipliers, e.g. 1.00GiB. In fact multipliers already aren't used for
figures less than 1024.
[PATCH] btrfs-progs: use IEC units for size
http://permalink.gmane.org/gmane.comp.file-systems.btrfs/26888https://patchwork.kernel.org/patch/2825841/
Make GParted capable of also accepting IEC prefix multipliers, just "B"
for bytes and no multiplier, as well as an optional space between the
number and multiplier. Therefore accept values like these:
1.00GB 1.00 GB
1.00GiB 1.00 GiB
1073741824B 1073741824 B
1073741824
Closes Bug #706914 - Prepare for btrfs tools using IEC prefix
multipliers
With recent btrfs-progs, GParted failed to format a btrfs file system
over the top of an existing one. Make btrfs failed with this error:
# mkfs.btrfs /dev/sdb1
...
/dev/sdb1 appears to contain an existing filesystem (btrfs).
Use the -f option to force overwrite.
With this commit to btrfs-progs on 2013-02-19, mkfs.btrfs checks for
existing file system signatures, including all mirror copies of btrfs
super blocks, before writing to the partition.
http://git.kernel.org/cgit/linux/kernel/git/mason/btrfs-progs.git/commit/?id=2a2d8e1962e8b6cda7b0a7584f6d2fb95d442cb6
btrfs-progs: require mkfs -f force option to overwrite filesystem or partition table
Make GParted clear all the mirror copies of the btrfs file system super
blocks as erase_filesystem_signatures() is intended to prevent detection
of old signatures. This also avoids having to determine if the -f
option to mkfs.btrfs is available before trying to use it.
Closes Bug #705426 - Formatting Existing BTRFS Partition as BTRFS Fails
Because mkfs.btrfs Is Not Run with "-f"
Unfortunately a regression was introduced with GParted 0.14.1 wherein
linux-swap grow and shrink operations were broken. This regression was
a direct result of the following commit:
Prevent file system grow when partition grow fails (#686668)
fd96328920
The regression did not result in an error in the GUI, but rather the
linux-swap file system was not "resized" to the correct partition size.
The logic error introduced with the above commit listed has now been
fixed.
BACKGROUND INFORMATION ON LINUX-SWAP RESIZING
Currently no command line tools are available for resizing linux-swap.
However, since linux-swap does not contain data in the usual sense, we
simulate resizing linux-swap by re-creating (maximizing to fill
partition) the linux-swap file system with the same UUID and volume
label. To work correctly, the linux-swap file system must be re-created
after the final partition resize action has occurred.
Closes Bug #706604 - Failure to properly grow or shrink linux-swap when
resizing
Under GNOME Shell a running instance of GParted was getting named
"Gpartedbin". Ultimately GNOME Shell [1] uses the WM_CLASS X Window
property [2], which GDK defaults from the executable name.
$ xprop WM_NAME WM_CLASS
(Click on running GParted window)
WM_NAME(STRING) = "/dev/sda - GParted"
WM_CLASS(STRING) = "gpartedbin", "Gpartedbin"
Set the Class name, second string of the WM_CLASS X Window property, to
"GParted". (This prevents the use of the '--class' GTK+/GDK common
command line option [3] to override the class name, but I expect nobody
ever uses the option. GNOME Terminal has the same limitation. It also
has the same change setting the Class name [4]).
References:
[1] Application Based GNOME 3
https://wiki.gnome.org/GnomeShell/ApplicationBased
[2] WM_CLASS Property, Inter-Client Communication Conventions Manual
http://tronche.com/gui/x/icccm/sec-4.html#s-4.1.2.5
[3] Running GTK+ Applications, Common command line options
https://developer.gnome.org/gtk3/stable/gtk-running.html
[4] Bug #685742 - Window class of terminals doesn't match the desktop
file name
https://bugzilla.gnome.org/show_bug.cgi?id=685742https://git.gnome.org/browse/gnome-terminal/commit/?id=3370c0e51159f5be476b909e94ac05e5362dd28a
Closes Bug #705323 - Shows up as 'Gpartedbin' in GNOME Shell
Dosfstools >= 3.0.18, released June 2013, renamed the programs thus:
dosfslabel becomes fatlabel,
dosfsck becomes fsck.fat,
and mkdosfs becomes mkfs.fat.
Dosfstools creates symbolic links for the old names for backward
compatibility, but unfortunately the Debian dosfstools-3.0.22-1
(experimental) package doesn't include those symbolic links. This
causes create, check and read unmounted FAT16/32 file systems to not be
supported.
Make GParted look for the new names first and the old names second.
Closes Bug #704629 - Program name changes in dosfstools 3.0.18+ break
FAT16/32 support
It was difficult to retrieve whether a filesystem's label can be set on reformat.
The read_label flag can't be used as it decides whether to use the logic in the filesystem class
rather than the fallback in GParted::set_device_partitions, to determine the label of a partition.
The create_with_label flag is NONE for file systems that we cannot format with a
label (or that we cannot format at all).
The value is usually EXTERNAL for file systems that we can format with a label.
The read-only functionality is unused and the readonly parameter is
always false in copy_filesystem() and copy_blocks() methods. This has
been the case since the copy simulation was dropped by commit:
b9b4b2e55d
Remove simulation pass ( read test ) on move
The minimum NTFS volume size supported by the ntfs-3g mkntfs command
is 1 MiB. Since 1 extra sector is required for the backup NTFS boot
sector, an NTFS partition must be at least 1 MiB plus 1 sector.
To demonstrate, create a 1 MiB unformatted partition using MiB
alignment (/dev/sdd1), and then try to format the partition as NTFS.
# mkntfs -Q -v -L "ntfs-test" /dev/sdd1
Device is too small (1023kiB). Minimum NTFS volume size is 1MiB.
For GParted this means a minimum NTFS partition size of 2 MiB since
the smallest unit in the GUI is MiB.
Bug #697848 - Failure to Create 1 MiB NTFS Partition
Place the size and noaction options before the device name when resizing
an NTFS file system to match the order documented in the manual page.
Example fixed command:
# ntfsresize -P --force --force -s 104857600 --no-action /dev/sda11
Ext[234] file systems were being shrunk 1 KiB too small. Ntfs and
reiserfs file systems were being shrunk 1 byte too small. The resultant
file systems were ending up 1 block smaller than they could have been.
This looks like an accidental leftover from when GParted use to:
1) Shrink the file system 1 cylinder too small
2) Shrink the partition
3) Grow the file system to fill the partition
Relevant commit:
d663c3c277
removed cylindersize buffering during resize from the filesystems
Closes Bug #701075 - Setting the backup boot sector when resizing NTFS
It was possible to make GParted crash by adding a label, check or new
UUID operation and then applying the operation before the view of
pending operations had finished fully opening. The operation would be
successfully applied but GParted would crash afterwards.
The fault was that Add_Operation() still enabled the Undo and Apply
buttons and processed the GTK event loop before merging the list of
pending operations. Faulty code flow went like this:
activate_*()
Add_Operation()
Add operation to the operations[] vector
Enable Undo and Apply buttons
Refresh_Visual()
Process GTK event loop
Process Apply button callback applying operations,
refreshing display and clearing operations[] vector
Merge operations in the operations[] vector
<< Core dump here >>
Merge_Operations()
Refresh_Visual()
This faulty code flow came about when merging of operations was added
and it didn't appreciate that the operations[] vector could have been
processed and cleared by Add_Operations() before the merge step.
Relevant commit:
b10349ae37
Merge overlapping operations (#438573)
Fragment of code in the label operation case:
2454 void Win_GParted::activate_label_partition()
2455 {
...
2472 Add_Operation( operation ) ;
2473
2474 // Verify if the two operations can be merged
2475 for ( unsigned int t = 0 ; t < operations .size() - 1 ; t++ )
2476 {
2477 if ( operations[ t ] ->type == OPERATION_LABEL_PARTITION )
2478 {
2479 if ( Merge_Operations( t, operations .size() - 1 ) )
2480 break;
2481 }
2482 }
Commentary in the crashing label operation case:
2472 The pending operation was already applied when Add_Operation()
returned resulting in the operations[] vector being cleared
setting its size to 0.
2475 The return type of operations.size() is an unsigned integral, so
the upper limit of the for loop is t < 0UL - 1. Assuming a
32-bit machine that's t < 4294967295.
2477 operations[] vector is access from out of bounds offset 0
upwards until unallocated memory is accessed resulting in a core
dump.
Fix this by not enabling the Undo and Apply buttons and processing the
GTK event loop until after merging of operations has been performed.
Fixed code flow goes like this:
activate_*()
Add_Operation()
Add operation to the operations[] vector
Merge operations in the operations[] vector
Merge_Operations()
show_operationslist()
Enable Undo and Apply buttons
Refresh_Visual()
Process GTK event loop
Process Apply button callback applying operations,
refreshing display and clearing operations[] vector
Not allowing the operations list to be process until after the merge
step is the be correct ordering. This also prevents the new operation
from flashing up in the operations list and then immediately
disappearing if merged. In the case of adding the first operation,
delaying enabling the Undo and Apply buttons is enough as the buttons
were previously disabled preventing the operation being applied before
the merge. In the case of adding further operations, processing of the
GTK event loop must also be delayed until after the merge to prevent the
operations being applied before the merge. Although that window of
opportunity would only be microseconds.
Bug #699452 - Crash when applying operations before pending operations
fully displayed
Mlabel sometimes writes uninitialised memory at the end of the label.
This causes mlabel, and therefore GParted, to display extra junk at the
end of the label. Depending on the bytes written GParted may also show
the following error on stdout:
(gpartedbin:18116): glibmm-CRITICAL **:
unhandled exception (type Glib::Error) in signal handler:
domain: g_convert_error
code : 1
what : Invalid byte sequence in conversion input
This is caused by a bug in mlabel, believed fixed in mtools 4.0.14.
Effects at least Fedora 14, RHEL/CentOS 6.x and Debian 6. (Use label
"1234567890" on Debian 6 to reproduce). Reproduction steps:
# mkdosfs -F16 /dev/sda7
mkdosfs 3.0.9 (31 Jan 2010)
# export MTOOLS_SKIP_CHECK=1
# mlabel ::123456 -i /dev/sda7
# mlabel -s :: -i /dev/sda7
Volume label is 123456~1t
It is not possible to detect which characters are junk so they can't be
trimmed. Instead just space pad labels so that at least newly written
labels aren't effected. (Fat labels are space padded on the disk by
definition anyway).
Bug #700228 - FAT16/32 labels are sometimes shown corrupted
There was virtually no difference between the separate modules for fat16
and fat32. Remove module fat32 and patch fat16 to serve both file
system subtypes. This is equivalent to what was previously done for
ext[234] by commit:
38dc55d49c
Combine duplicate code for ext[234]
Rename function to update_command_output() to better reflect that this
callback updates the UI with the latest output read from the command
being executed.
This makes more sense knowing a future change will add a separate
callback which parses the output read from the command and updates the
progress bar. This function should be named update_command_progress().
Rename the libsigc++ signals to signal_update and signal_eof to match
the naming used for signals in GParted.
fgrep 'sigc::signal' include/*.h
Also explicitly use the emit() method rather than using the object
operator(). This again is to match the convention in GParted and make
it more obvious what is happening.
fgrep '.emit(' include/*.h
The previous commit missed one glibmm GSource wrapper in the form of the
io watch for the PipeCapture class. Convert this one to use glib
directly as well.
Bug #697727 - Segfault in livecd Gparted v 0.15.0-3 when copying
partition
On RHEL / CentOS 5.9 GParted couldn't set an icon as the
set_default_icon_name() method is not available. See commit [1] for
details.
Re-add the old set_icon_from_file() method as a fallback and re-install
a GParted pixmap as was used before commit [2].
Commit [1]:
a042107883
Only use Gtk::Window::set_default_icon_name method when available
Commit [2]:
f5a80bc904
Enabled GParted to use themed app icon (Tango theme)
Bug #695279 - GParted doesn't compile on RHEL / CentOS 5.9
On RHEL / CentOS 5.9 GParted failed to run any external commands and
instead reported the following warnings to the terminal:
# src/gpartedbin
======================
libparted : 1.8.1
======================
Failed to change to directory '' (No such file or directory)
Failed to change to directory '' (No such file or directory)
Failed to change to directory '' (No such file or directory)
...
Utils::execute_command() and FileSystem::execute_command() passed a zero
length string for the working directory to
Glib::spawn_async_with_pipes() to mean don't change directory. Instead
glibmm just tried to change to the directory with a zero length name.
This was fixed with glibmm >= 2.21.2 released July 2009, however RHEL /
CentOS 5.9 only has glibmm 2.12.10.
Relevant glibmm fix:
Treat empty Glib::spawn*() working dir as unset
https://git.gnome.org/browse/glibmm/commit/?id=8a7805cbbe6d268e975669349beb4e82d971537d
Fix by simply specifying ".", the current working directory, as the
directory to change into before executing all commands.
Bug #695279 - GParted doesn't compile on RHEL / CentOS 5.9
Glib::ustring::compose() method requires glibmm >= 2.16, but RHEL /
CentOS 5.9 only provides glibmm 2.12. Replace with String::ucompose()
as is used everywhere else in the code.
Add missing include for kill() and SIGINT declarations.
Bug #695279 - GParted doesn't compile on RHEL / CentOS 5.9
Roughly HFSX is a case sensitive version of the HFS+ file system.
Parted reports such a file system as "hfsx" rather than "hfs+".
# mkfs.hfsplus -v "case insensitive hfs+" /dev/sda7
Initialized /dev/sda7 as a 1024 MB HFS Plus volume
# mkfs.hfsplus -s -v "case sensitive hfs+" /dev/sda8
Initialized /dev/sda8 as a 1024 MB HFS Plus volume
# parted /dev/sda print
...
Number Start End Size Type File system Flags
...
7 356GB 357GB 1074MB logical hfs+
8 357GB 358GB 1074MB logical hfsx
# blkid /dev/sda[78]
/dev/sda7: LABEL="case insensitive hfs+" TYPE="hfsplus"
/dev/sda8: LABEL="case sensitive hfs+" TYPE="hfsplus"
Make GParted recognise HFSX file system variants too.
Closes Bug #698876 - GParted fails to recognize HFS+ partition (possible
due to disabled journaling)
In some circumstances gparted would appear to scan forever if it
encountered a blank hard disk. This would happen if the timing of
events was right and gparted encountered a disk without a partition
table. The missing partition table would cause a call to the
exception handler, which could get stuck in a thread waiting position.
The problem was traced back to the ped_exception_handler
and a cond.signal() call missing the requisite mutex.lock() and
mutex.unlock() around the signal call.
Closes Bug #697518 - gparted scans forever blank disk in virtualbox
The glibmm GSource wrappers have a bug where they do not do
reference counting properly, and have a race condition where
the background thread can try to touch the source after the
main thread has already processed and destroyed it. This
results in writes to freed memory and sometimes this causes
crashes or other erratic behavior. Avoid using the glibmm
wrappers and use glib directly. See bug #561885 for details
of the glibmm bug.
Bug #697727 - Segfault in livecd Gparted v 0.15.0-3 when copying partition
For an absolutely full NTFS GParted doesn't show the file system usage
but instead shows a warning against the partition.
GParted was looking only for "resize at %d" to determine file system
used figure in the output from the ntfsresize command. Also handle
finding "ERROR: Volume is full" too to derive used figure.
NTFS with 1 block free case:
# ntfsresize --info --force --no-progress-bar /dev/sda9
...
Current volume size: 1073738240 bytes (1074 MB)
...
Collecting resizing constraints ...
You might resize at 1073737728 bytes (freeing 4096 bytes).
...
# echo $?
0
Absolutely full NTFS case:
# ntfsresize --info --force --no-progress-bar /dev/sda9
...
Current volume size: 1073738240 bytes (1074 MB)
...
Collecting resizing constraints ...
ERROR: Volume is full. To shrink it, delete unused files.
# echo $?
1
Closes Bug #697946 - Absolutely full NTFS reported as partition error
Lookup the maximum reiser4 file system label length instead, so that it
is defined in only one place. Small improvement missed from earlier
commit:
18941e24d3
Avoid reading trailing junk for a reiser4 label (#689318)
These functions in GParted_Core:
open_device()
open_device_and_disk()
close_disk()
close_device_and_disk()
call the following functions in the libparted API:
ped_device_get()
ped_disk_new()
ped_disk_destroy()
ped_device_destroy()
which don't open or close anything. Instead they allocate and
deallocate PedDevice and PedDisk memory structures which describe block
devices and partition tables respectively.
Rename functions:
open_device_and_disk() -> get_device_and_disk()
close_device_and_disk() -> destroy_device_and_disk()
and merge open_device() and open_device() as each only wrapped one
libparted function and was only called from a single place.
The wipefs command has the following significant limitations which were
worked around in previous commits:
1) Wasn't available in the earliest distributions supported by GParted;
2) Had to be called 3 times to erase vfat (fat16/32) signatures in all
but the most recent versions.
This meant we had all the code to clear file system signatures without
using the wipefs command as well as extra complexity of using wipefs
too. So just remove use of the wipefs command.
Bug #688882 - Improve clearing of file system signatures
Wipefs only clears the nilfs2 file system signature from the primary
super block at the start of the partition. This is enough for blkid to
no longer detect the file system. However parted (>= 2.4 with nilfs2
support) and therefore GParted still detect the file system using the
secondary super block at the end of the partition.
# mkfs.nilfs2 /dev/sda12
# wipefs -a /dev/sda12
2 bytes were erased at offset 0x00000406 (nilfs2): 34 34
# blkid /dev/sda12
# parted /dev/sda12 print
Model: Unknown (unknown)
Disk /dev/sda12: 1074MB
Sector size (logical/physical): 512B/512B
Partition Table: loop
Disk Flags:
Number Start End Size File system Flags
1 0.00B 1074MB 1074MB nilfs2
Overwrite the nilfs2 secondary super block at the end of the partition
with 4K block of zeros. Always do this just in case the partition
contains a nilfs secondary super block but is detected as a different
file system. Such an example was documented in an earlier commit
message.
Bug #688882 - Improve clearing of file system signatures
As failure of flushing the cache can lead to misidentification of file
systems make it a checked step in the displayed operation details.
Bug #688882 - Improve clearing of file system signatures
When wipefs command fails this means that in the display of the
operation results tree:
1) The command is displayed with a time and a stop sign, rather than no
time and a warning sign;
2) There is no report of any warnings or failures at the top level
summary.
(The immedately following "clear primary signatures" step will clear all
the file system signatures which wipefs failed to do. Should this fail
too it will report errors which will fail the whole operation).
Bug #688882 - Improve clearing of file system signatures
RHEL/CentOS 5.x uses util-linux 2.13 which predates the wipefs command.
Also the wipefs command may fail. Therefore implement a fallback which
caters for both these cases by overwriting all possible file system
super blocks with zeros at the start of the partition.
The "MUST be cleared" requirement stated in the first patch is now
satisfied. Therefore the status of the "clear old file system
signatures" step, as returned by erase_filesystem_signatures(), now
reflects the success of the wipefs command or the internal write zeros.
Failure of both will stop all operations, matching what happens with all
other steps.
Bug #688882 - Improve clearing of file system signatures
Before util-linux 2.21.0, released Feb 2012, wipefs only cleared one of
the three vfat (fat16/fat32) signatures it can be detected by each time
wipefs was run. Also if a nilfs2 file system was created before all
three signatures were cleared the partition was still recognised as a
vfat file system, albeit a corrupted one, rather than as a nilfs2 file
system.
Old wipefs clearing vfat signatures:
# wipefs --version
wipefs from util-linux 2.20.1
# wipefs -a /dev/sda7
8 bytes were erased at offset 0x52 (vfat)
they were: 46 41 54 33 32 20 20 20
# wipefs -a /dev/sda7
1 bytes were erased at offset 0x0 (vfat)
they were: eb
# wipefs -a /dev/sda7
2 bytes were erased at offset 0x1fe (vfat)
they were: 55 aa
New wipefs clearing vfat signatures:
# wipefs --version
wipefs from util-linux 2.21.2
# wipefs -a /dev/sda12
8 bytes were erased at offset 0x00000052 (vfat): 46 41 54 33 32 20 20 20
1 bytes were erased at offset 0x00000000 (vfat): eb
2 bytes were erased at offset 0x000001fe (vfat): 55 aa
Workaround by calling "wipefs -a" three times if the output indicated
only one vfat signature was cleared.
Bug #688882 - Improve clearing of file system signatures
When just formatting an existing partition to "cleared", GParted just
uses wipefs to clear the file system signatures. Afterwards parted/
libparted still detect the file system and GParted shows errors from the
file system specific tools reporting the file system doesn't exist.
# wipefs /dev/sda7
offset type
----------------------------------------------------------------
0x0 xfs [filesystem]
UUID: 28399a74-83a5-4ed7-aaf8-c76ac449fb57
# wipefs -a /dev/sda7
4 bytes were erased at offset 0x0 (xfs)
they were: 58 46 53 42
# parted /dev/sda print
Model: ATA SAMSUNG HM500JI (scsi)
Disk /dev/sda: 500GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
1 1049kB 538MB 537MB primary ext4 boot
2 538MB 5907MB 5369MB primary linux-swap(v1)
3 5907MB 32.8GB 26.8GB primary ext4
4 32.8GB 500GB 467GB extended
5 32.8GB 355GB 322GB logical ext3
6 355GB 356GB 1074MB logical
7 356GB 357GB 1074MB logical xfs
# xfs_db -c 'sb 0' -c 'print blocksize' -c 'print dblocks' -c 'print fdblocks' -r /dev/sda7
xfs_db: /dev/sda7 is not a valid XFS filesystem (unexpected SB magic number 0x00000000)
Wipefs was run on the partition specific block device (/dev/sda7) where
as libparted reads the disk using the whole disk device (/dev/sda).
However as the Linux buffer cache does not provide cache coherency, the
xfs file system can still be found in the cache of /dev/sda.
Fix this by calling ped_device_sync() after wipefs as it guarantees
cache coherency.
(As documented in erase_filesystem_signatures() there are cases when
calling ped_device_sync() isn't necessary, but testing shows that the
whole processes takes at most 0.15 seconds. As this is in the middle of
applying an operation, uses won't notice this extra time so just always
call ped_device_sync()).
Bug #688882 - Improve clearing of file system signatures
Add "cleared" to the bottom of list of file system formats available in
the Create new Partition dialog and in the Format to --> (file system
list) menu. This clears existing file system signatures in the newly
created partitions and existing partitions respectively.
Bug #688882 - Improve clearing of file system signatures
Move some code into new create_format_menu_add_item() sub-function which
adds one file system entry to the Partition --> Format to -->
(file system list) menu.
Bug #688882 - Improve clearing of file system signatures
Previously the function erase_filesystem_signatures() was used to clear
file system signatures when a new partition was created and when an
existing partition was formatted with a file system. However this was
only available with libparted <= 2.4 and then only for the file systems
which libparted supports.
Having multiple different file system signatures on a partition leads to
misidentification of file system. For example creating a nilfs2 over
the top of a fat32 file system is detected as a fat32, not nilfs2. This
shows that old file system signatures must be cleared before a new file
system is created.
Fix by always using "wipefs -a /dev/PARTITION" command to clear all old
file system signatures rather than libparted API calls. Failure from
wipefs is only considered a warning so doesn't fail the file system
creation. (This doesn't yet fully meet the "MUST be cleared"
requirement above. Will be fully met later in this patchset). Output
from the wipefs command is displayed as a new sub-step which looks like
this:
v Format /dev/sda7 as xfs 00:00:05
> calibrate /dev/sda14 00:00:01
v clear old file system signatures in /dev/sda7 00:00:01 [NEW]
> wipefs -a /dev/sda7 [NEW]
> set partition type on /dev/sda7 00:00:02
v create new xfs file system 00:00:01
> mkfs.xfs -f -L "" /dev/sda7
Also signatures are only cleared immediately before a new file system is
written and not when an unformatted partition is created. This allows
recovery from accidental partition deletion by re-creating the deleted
partition as unformatted.
Bug #688882 - Improve clearing of file system signatures
GParted fails to compile on RHEL/CentOS 5.9 because it doesn't provide
the Gtk::Window::set_default_icon_name() method. The "gtkmm GTK::Window
Class Reference" document says that set_default_icon_name() is available
in gtkmm >= 2.6, however it is not available in RHEL/CentOS 5.9 with
gtkmm 2.10.
Add an autoconf compile and link check for the set_default_icon_name()
method and make GParted only use the method when available.
Bug #695279 - GParted doesn't compile on RHEL / CentOS 5.9
GParted fails to compile on RHEL/CentOS 5.9 because it doesn't provide
the Glib::Regex class. Glib::Regex class requires glibmm >= 2.14,
however RHEL/CentOS 5.9 only provides glibmm 2.12.
Add an autoconf check for the Glib::Regex class and fallback code using
the POSIX regex function. Fall back code is the same as that used prior
to commit:
b6f1c56fb1
Enhance regexp_label method to handle unicode characters
Bug #695279 - GParted doesn't compile on RHEL / CentOS 5.9
Only supports detection and creation of f2fs file systems. Requires
f2fs-tools and a blkid with f2fs support, util-linux > 2.22.2.
f2fs-tools v1.1.0 only supports file system creation.
Currently requires util-linux directly from the git repository as f2fs
support was only committed on 5 Feb 2013 and it has not yet been
released.
Closes Bug #695396 - Please apply f2fs patch
After a refresh, instead of having no partition selected, default to the
largest unallocated partition. This allows immediate creation of a new
partition in the unallocated space.
Bug 667365 - Free space should be selected by default
Many filesystems do not implement some of their methods, but had to provide
dummy implementations. Remove all of the dummy implementations and instead
just provide one in the base FileSystem class.
The details view refused to use additional space, even after the window was
expanded, instead continuing to use the scrollbars. Now resizing the
window will be allowed regardless of the state of the details expander, and
the details view will expand to use the extra space. Also request enough
initial width to not need a horizontal scrollbar.
Closes:
Bug 602635 - list of tasks in apply dialog does not expand to the available
vertical space
Bug 662722 - Increase default width of "applying..." dialog to include the
"Details" status icons
There were separate modules for ext3 and ext4 even though there
were virtually no differences with ext2. Remove the duplicate
modules and patch ext2 to serve as a common reference for all
three sub types.
After a move is canceled or fails, an attempt is made to revert the disk to
its previous state. This was moving back all blocks that had already been
copied. Many of the first blocks copied during a move do not overwrite
any valid data, and so do not need put back after an error, so don't bother
doing so.
After a failed or canceled move, the move was reverted, and the filesystem
was fscked while the partition was still in the expanded state encompassing
the original and destination partitions, resulting in errors and
corruption. Move fsck to after the partition table has been restored to
its original state.
After a refresh, instead of having no partition selected, default to the
largest unallocated partition. This allows immediate creation of a new
partition in the unallocated space.
Closes Bug 667365 - Free space should be selected by default
Interested operations can now connect a signal to their OperationDetail
to be notified of a cancelation request. The internal copy/move code
will now cleanly stop on cancelation, allowing the partition to be
rolled back to its previous state. This makes canceling a move
perfectly safe.
After clicking cancel, the button changes to "Force Cancel" and is
disabled for 5 seconds. Operations that are safe to cancel will do so
and those that are not will continue to run. Clicking force cancel
asks operations to cancel, even if doing so is unsafe. For the
internal copy/move algorithm, canceling is always safe because an
error results in a rollback operation. Canceling the rollback is
unsafe. For external commands, filesystem modules may indicate
that the command is safe to cancel or not. Canceled commands will
be terminated with SIGINT.
As a result of the new safe cancel vs force cancel distinction, the
scary warning about cancl causing corruption has been moved to
after clicking the force cancel button.
Part of Bug #601239 - Please allow 'Cancel after current operation'
Have the copy code create a background thread to do the actual copying so
that it won't block the main loop.
Part of Bug 685740 - Refactor to use asynchronous command execution
Win_Gparted and Dialog_Progress were creating threads to perform most
functions in the background. Most of the time, the only reason the
threads blocked was to execute an external command. The external command
execution has been changed to spawn the command asynchronously and wait
for completion with a nested main loop. While waiting for completion,
the pipe output is captured via events. In the future, this will allow
for it to be parsed in real time to obtain progress information.
Those tasks in GParted_Core that still block now spawn a background thread
and wait for it to complete with a nested main loop to avoid hanging the
gui.
Part of Bug #685740 - Refactor to use asynchronous command execution
fat16 and fat32 were creating a temp mtoolsrc file to configure the
command to reference a drive letter and ignore certain errors. They have
been changed to pass this information via the command line and environment
instead.
Dialog_Progress was using pthread_create() so that it could later
pthread_cancel() the thread. pthread_cancel() is wildly unsafe and full
of errors. Changed to use Glib's threads like the rest, and only cancel
between operations. Because it can take some time to cancel, disable
the cancel button once it has been clicked once.
Bug 601239 - Please allow 'Cancel after current operation'
Win_Gparted and Dialog_Progress were looping on Gtk::Main::events_pending()
and iteration() with usleeps in between. Use a full mainloop instead and
a proper timeout to trigger pulsebar updates instead of usleeps.
Part of Bug 685740 - Refactor to use asynchronous command execution
Use of these functions is depreciated and making gtk calls in a background
thread still sometimes causes deadlocks or crashes. Change ped exception
handler to instead use an idle function to force the main thread to display
the dialog box.
Part of Bug 685740 - Refactor to use asynchronous command execution
When moving a MiB aligned primary partition from right to left, the
resulting partition was unexpectedly 1 MiB smaller in size.
From further testing, this occurred only if the move boundaries
overlapped the original partition boundaries.
In cases where the move did not overlap the original partition
boundaries, then the size remained the same.
Closes Bug #695078 - Move of MiB aligned partition right to left
yields unexpected shrink of 1 MiB
Add -I option to 'mkdosfs' command to work around faulty detection of
"complete disks" vs. "partitions".
Bug #693955 - mkdosfs detects "complete disk" vs. "partition" incorrectly
Thanks to Kano for reporting this!
(This will also be needed if GParted ever allows formatting a device without
partitioning it.)
No longer need to trim fat16, fat32 and xfs labels as all labels are
limited to their maximum lengths during entry.
Bug #689318 - filesystem type specific support for partition name
maximum length
For a reiser4 file system with exactly the maximum size 16 character
label, debugfs.reiser4 may include junk at the end of the label it
prints. (It reads the label into a 16 character array and prints it as
a string, but there isn't a nul terminating character).
# mkfs.reiser4 --yes --label abcdefghij123456 /dev/sda13
# debugfs.reiser4 /dev/sda13 2> /dev/null | grep label:
label: abcdefghij123456!
Trim the read label to at most 16 characters.
Bug #689318 - filesystem type specific support for partition name
maximum length
Active Linux software RAID devices are detected in the
Proc_Partitions_Info method. Hence the SWRaid method is no longer
required.
Removal of the SWRaid method fixes the problem with the error message:
Could not stat device /dev/md/0 - No such file or directory
This fixes the problem because we no longer use "mdadm --examine
--scan" in an attempt to detect Linux software RAID devices. The
mdadm command was returning device names such as /dev/md/0, which are
incorrect for GParted.
NOTE: With this change, GParted no longer requires the mdadm command
to detect Linux software RAID devices.
Closes Bug #678379 - Could not stat device /dev/md/0 - No such file or
directory
Add regular expression to detect the device entry only (e.g., md127)
and not the partitions (e.g., md127p1).
Linux software RAID devices are listed in /proc/partitions as follows:
# cat /proc/partitions
major minor #blocks name
9 127 523968 md127
259 0 131072 md127p1
259 1 262144 md127p2
Part of Bug #678379 - Could not stat device /dev/md/0 - No such file
or directory
HP Smart Array Devices are listed in /proc/partitions as follows:
major minor #blocks name
104 0 143338560 cciss/c0d0
104 1 104391 cciss/c0d0p1
104 2 24097 cciss/c0d0p2
104 3 143203410 cciss/c0d0p3
The previous regular expression was too broad, and would match many
other forms, including Linux software RAID devices such as md127.
Linux software RAID devices are listed in /proc/partitions as follows:
major minor #blocks name
9 127 523968 md127
259 0 131072 md127p1
259 1 262144 md127p2
This problem with the regexp was discovered while investigating a
problem with Linux software RAID device detection.
The regular expression has been tightened up to match the following
hardware RAID controllers:
Compaq/HP Smartarray RAID controller
E.g., device = /dev/cciss/c0d0, partition = /dev/cciss/c0d0p1
(linux-x.y.z/Documentation/blockdev/cciss.txt)
Compaq SMART2 Intelligent Disk Array controller
E.g., device = /dev/ida/c0d0, partition = /dev/ida/c0d0p1
(linux-x.y.z/Documentation/blockdev/cpqarray.txt)
Mylex DAC960/AcceleRAID/eXtremeRAID PCI RAID Controllers
E.g., device = /dev/rd/c0d0, partition = /dev/rd/c0d0p1
(linux-x.y.z/Documentation/blockdev/README.DAC960)
Related to Bug #678379 - Could not stat device /dev/md/0 - No such
file or directory
In certain situations with a mixture of partition alignments, the MiB
alignment option would try to set the end of a logical partition
beyond the end of the extended partition. This, of course, is an
invalid partition geometry and the libparted library correctly reports
failure when this is attempted.
This enhancement to MiB alignment adds a check to see if the end of
the logical partition would be beyond the extended partition, and
adjusts the end of the logical partition if required.
Closes Bug #686668 - Growing logical partition overlaps end of
extended partition
Prior to this enhancement, GParted ignored the status of the partition
grow operation, and attempted to grow the file system.
Now we check the status of the partition grow operation prior to
growing the file system.
Part of Bug #686668 - Growing logical partition overlaps end of
extended partition
Mike Fleetwood