This function is an extended version of `chmod(2)` that lets one control
whether to dereference symlinks, and specify a file descriptor to a
directory that will be used as the base for relative paths.
Previously we were uncaching inodes from TmpFSInode::one_ref_left().
This was not safe, since one_ref_left() was effectively being called
on a raw pointer after decrementing the local ref count and observing
it become 1. There was a race here where someone else could trigger
the destructor by unreffing to 0 before one_ref_left() got called,
causing us to call one_ref_left() on a deleted inode.
We fix this by using the new remove_from_secondary_lists() mechanism
in ListedRefCounted and synchronizing all access to the TmpFS inode
map with the main Inode::all_instances() lock.
There's probably a nicer way to solve this.
When writing into a block via an O_DIRECT open file description, we
would first flush every dirty block *except* the one we're about to
write into.
The purpose of flushing is to ensure coherency when mixing direct and
indirect accesses to the same file. This patch fixes the issue by only
flushing the affected block.
Our behavior was totally backwards and could end up doing a lot of
unnecessary work while avoiding the work that actually mattered.
Since this function always returns a CacheEntry& (after potentially
evicting someone else to make room), let's call it "ensure" instead of
"get" to match how we usually use these terms.
Previously we were using this vector to store the inodes as we iterated.
However, we don't need to store all of them, just the previous inode, as
we know it will be safe to remove it once we've iterated past that
element.
This modifies sys$chown to allow specifying whether or not to follow
symlinks and in which directory.
This was then used to implement lchown and fchownat in LibC and LibCore.
This is required for SlavePTY's custom unref handler to function
correctly, as otherwise a SlavePTY held in a File RefPtr would call
the base's (RefCounted<>) unref method instead of SlavePTY's version.
Since RefCounted automatically calls a method named `will_be_destoyed`
on classes that have one, so there's no need to have a custom
implementation of unref in File.
This will allow File and it's descendants to use RefCounted instead of
having a custom implementation of unref. (Since RefCounted calls
will_be_destroyed automatically)
This commit also removes an erroneous call to `before_removing` in
AHCIPort, this is a duplicate call, as the only reference to the device
is immediately dropped following the call, which in turns calls
`before_removing` via File::unref.
Custody's unref is one of many implementions of ListedRefCounted's
behaviour in the Kernel, which results in avoidable bugs caused by
the fragmentation of the implementations. This commit starts the work
of replacing all custom implementations with ListedRefCounted by
porting Custody to it.
This was a premature optimization from the early days of SerenityOS.
The eternal heap was a simple bump pointer allocator over a static
byte array. My original idea was to avoid heap fragmentation and improve
data locality, but both ideas were rooted in cargo culting, not data.
We would reserve 4 MiB at boot and only ended up using ~256 KiB, wasting
the rest.
This patch replaces all kmalloc_eternal() usage by regular kmalloc().
We were doing this dance in notify_watchers():
set_metadata_dirty(true);
set_metadata_dirty(false);
This was done in order to force out inode watcher events immediately.
Unfortunately, this was racy, as if SyncTask got scheduled at the wrong
moment, it would try to flush metadata for a clean inode. This then got
trapped by the VERIFY() statement in Inode::sync_all():
VERIFY(inode.is_metadata_dirty());
This patch fixes the issue by replacing notify_watchers() with lazy
metadata notifications like all other filesystems.
If the data passed to sys$write() is backed by a not-yet-paged-in inode
mapping, we could end up in a situation where we get a page fault when
trying to copy data from userspace.
If that page fault handler tried reading from an inode that someone else
had locked while waiting for the disk cache lock, we'd deadlock.
This patch fixes the issue by copying the userspace data into a local
buffer before acquiring the disk cache lock. This is not ideal since it
incurs an extra copy, and I'm sure we can think of a better solution
eventually.
This was a frequent cause of startup deadlocks on x86_64 for me. :^)
As pointed out by BertalanD on Discord, POSIX specifies that
_SC_SYMLOOP_MAX (implemented in the following commit) always needs to be
equal or more than _POSIX_SYMLOOP_MAX (8, defined in
LibC/bits/posix1_lim.h), hence I've increased it to that value to
comply with the standard.
The move to header is required for the following commit - to make this
constant accessible outside of the VFS class, namely in sysconf.
This fixes at least half of our LibC includes in the kernel. The source
of truth for errno codes and their description strings now lives in
Kernel/API/POSIX/errno.h as an enumeration, which LibC includes.
Instead, allocate before constructing the object and pass NonnullOwnPtr
of KString to the object if needed. Some classes can determine their
names as they have a known attribute to look for or have a static name.
This file refers to the controlling terminal associated with the current
process. It's specified by POSIX, and is used by ports like openssh to
interface with the terminal even if the standard input/output is
redirected to somewhere else.
Our implementation leverages ProcFS's existing facilities to create
process-specific symbolic links. In our setup, `/dev/tty` is a symbolic
link to `/proc/self/tty`, which itself is a symlink to the appropriate
`/dev/pts` entry. If no TTY is attached, `/dev/tty` is left dangling.
This option is already enabled when building Lagom, so let's enable it
for the main build too. We will no longer be surprised by Lagom Clang
CI builds failing while everything compiles locally.
Furthermore, the stronger `-Wsuggest-override` warning is enabled in
this commit, which enforces the use of the `override` keyword in all
classes, not just those which already have some methods marked as
`override`. This works with both GCC and Clang.
This isn't a complete conversion to ErrorOr<void>, but a good chunk.
The end goal here is to propagate buffer allocation failures to the
caller, and allow the use of TRY() with formatting functions.
We have no way of writing changes to memory-mapped files back to disk,
and software relying on this functionality for output would fail
miserably. Let's just return ENOTSUP instead to allow callers to fall
back to standard file IO instead of silently discarding writes.
This makes the LLD port work, which uses memory-mapped files to write
its output by default.
Some calls of copy_to_user were converting Userspace<T*> to
Userspace<U*> via the implicit conversion to FlatPtr. Change them to use
the static_ptr_cast overload that is designed to express this conversion
... In files included from Kernel/Thread.cpp or Kernel/Process.cpp
Some places the warning is suppressed, because we do not want a const
object do have non-const access to the returned sub-object.
Instead of signalling allocation failure with a bool return value
(false), we now use ErrorOr<void> and return ENOMEM as appropriate.
This allows us to use TRY() and MUST() with Vector. :^)
We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace!
This was a slightly tedious refactoring that took a long time, so it's
not unlikely that some bugs crept in.
Nevertheless, it does pass basic functionality testing, and it's just
real nice to finally see the same pattern in all contexts. :^)
This FIXME does not seem to apply anymore. Yes, symbolic links in all
filesystems appear to be slightly broken, but that has nothing to do
with File::absolute_path. Let's remove the wrong FIXME instead of adding
to the confusion.
Found due to smelly code in InodeFile::absolute_path.
In particular, this replaces the following misleading methods:
File::absolute_path
This method *never* returns an actual path, and if called on an
InodeFile (which is impossible), it would VERIFY_NOT_REACHED().
OpenFileDescription::try_serialize_absolute_path
OpenFileDescription::absolute_path
These methods do not guarantee to return an actual path (just like the
other method), and just like Custody::absolute_path they do not
guarantee accuracy. In particular, just renaming the method made a
TOCTOU bug obvious.
The new method signatures use KResultOr, just like
try_serialize_absolute_path() already did.
Forcing the formatting to go through `Formatter<FormatString>` is
completely unnecessary, increases code size, performs a String
allocation and prevents us from using the formatting options available
on that type.
This commit also removes explicit formatters from
`BlockBasedFileSystem::BlockIndex` and `Kernel::InodeIndex`, as those
are already covered by the blanket implementation for all
`DistinctNumeric` types.
ProcFSGlobalInode now calls `write_bytes()`, `truncate()` and
`set_mtime()` on its associated component. This allows us to write 0 or
1 to a ProcFSSystemBoolean component to toggle a boolean value.
The platform independent Processor.h file includes the shared processor
code and includes the specific platform header file.
All references to the Arch/x86/Processor.h file have been replaced with
a reference to Arch/Processor.h.
Normally, trying to truncate a SysFSInode should result in EPERM error.
However, as suggested by Ali (@alimpfard), we can allow the PowerState
node to be "truncated" so one can open that file with O_TRUNC option.
Likewise, we also need to provide a way to set modified time on SysFS
inodes. For most inodes, we should return ENOTIMPL error, but for the
power state switch, we ignore the modified time setting and just return
KSuccess.
These fixes allow to do "echo -n 1 > /sys/firmware/power_state" in Shell
after gaining root permissions, to switch the power state.
Previously there was a mix of returning plain strings and returning
explicit string views using `operator ""sv`. This change switches them
all to standardized on `operator ""sv` as it avoids a call to strlen.
This singleton simplifies many aspects that we struggled with before:
1. There's no need to make derived classes of Device expose the
constructor as public anymore. The singleton is a friend of them, so he
can call the constructor. This solves the issue with try_create_device
helper neatly, hopefully for good.
2. Getting a reference of the NullDevice is now being done from this
singleton, which means that NullDevice no longer needs to use its own
singleton, and we can apply the try_create_device helper on it too :)
3. We can now defer registration completely after the Device constructor
which means the Device constructor is merely assigning the major and
minor numbers of the Device, and the try_create_device helper ensures it
calls the after_inserting method immediately after construction. This
creates a great opportunity to make registration more OOM-safe.
Because we were holding a strong ref to the OpenFileDescription in
LocalSocket and a strong ref to the LocalSocket in Inode, we were
creating a reference cycle in the event of the socket being cleaned up
after the file description did (i.e. unlinking the file before closing
the socket), because the file description never got destructed.
Instead of doing so in the constructor, let's do immediately after the
constructor, so we can safely pass a reference of a Device, so the
SysFSDeviceComponent constructor can use that object to identify whether
it's a block device or a character device.
This allows to us to not hold a device in SysFSDeviceComponent with a
RefPtr.
Also, we also call the before_removing method in both SlavePTY::unref
and File::unref, so because Device has that method being overrided, it
can ensure the device is removed always cleanly.
We are no longer have a separate Inode object class for the pts
directory. With a small exception to this, all chmod and chown code
is now at one place.
It's now possible to create any name of a sub-directory in the
filesystem.
The current implementation of DevFS resembles the linux devtmpfs, and
not the traditional DevFS, so let's rename it to better represent the
direction of the development in regard to this filesystem.
The abbreviation for DevTmpFS is still "dev", because it doesn't add
value as a commandline option to make it longer.
In quick summary - DevFS in unix OSes is simply a static filesystem, so
device nodes are generated and removed by the kernel code. DevTmpFS
is a "modern reinvention" of the DevFS, so it is much more like a TmpFS
in the sense that not only it's stored entirely in RAM, but the userland
is responsible to add and remove devices nodes as it sees fit, and no
kernel code is directly being involved to keep the filesystem in sync.
In order to make this kind of operation simpler, we no longer use a
Vector to store pointers to DevFSDeviceInode, but an IntrusiveList is
used instead. Also, we only allow to remove device nodes for now, but
in theory we can allow to remove all kinds of files from the DevFS.
These files are not marked as block devices or character devices so they
are not meant to be used as device nodes. The filenames are formatted to
the pattern "major:minor", but a Userland program need to call the parse
these format and inspect the the major and minor numbers and create the
real device nodes in /dev.
Later on, it might be a good idea to ensure we don't create new
SysFSComponents on the heap for each Device, but rather generate
them only when required (and preferably to not create a SysFSComponent
at all if possible).
Don't create these device nodes in the Kernel, so we essentially enforce
userspace (SystemServer) to take control of this operation and to decide
how to create these device nodes.
This makes the DevFS to resemble linux devtmpfs, and allows us to remove
a bunch of unneeded overriding implementations of device name creation
in the Kernel.
This patch adds KBufferBuilder::try_create() and treats it like anything
else that can fail. And so, failure to allocate the initial internal
buffer of the builder will now propagate an ENOMEM to the caller. :^)
- Renamed try_create_absolute_path() => try_serialize_absolute_path()
- Use KResultOr and TRY() to propagate errors
- Don't call this when it's only for debug logging
- Use KResultOr and TRY() to propagate errors
- Check for OOM errors
- Move allocation out of constructors
There's still a lot more to do here, as SysFS is still quite brittle
in the face of memory pressure.
The default template argument is only used in one place, and it
looks like it was probably just an oversight. The rest of the Kernel
code all uses u8 as the type. So lets make that the default and remove
the unused template argument, as there doesn't seem to be a reason to
allow the size to be customizable.
This commit moves the KResult and KResultOr objects to Kernel/API to
signify that they may now be freely used by userspace code at points
where a syscall-related error result is to be expected. It also exposes
KResult and KResultOr to the global namespace to make it nicer to use
for userspace code.
