Since the inode is the logical owner of its block list, let's move the
code that computes the block list there, and also stop hogging the FS
lock while we compute the block list, as there is no need for it.
There are two locks in the Ext2FS implementation:
* The FS lock (Ext2FS::m_lock)
This governs access to the superblock, block group descriptors,
and the block & inode bitmap blocks. It's held while allocating
or freeing blocks/inodes.
* The inode lock (Ext2FSInode::m_lock)
This governs access to the inode metadata, including the block
list, and to the content data as well. It's held while doing
basically anything with the inode.
Once an on-disk block/inode is allocated, it logically belongs
to the in-memory Inode object, so there's no need for the FS lock
to be taken while manipulating them, the inode lock is all you need.
This dramatically reduces the impact of disk I/O on path resolution
and various other things that look at individual inodes.
Since these filesystems operate on an underlying file descriptor
and rely on its offset for correctness, let's use the FS lock to
serialize these operations.
This also means that FS subclasses can rely on block-level read/write
operations being atomic.
This is basically just for consistency, it's quite strange to see
multiple AK container types next to each other, some with and some
without the namespace prefix - we're 'using AK::Foo;' a lot and should
leverage that. :^)
This reverts commit 1e737a5c50.
The cached block list does not include meta-blocks, so we'd end up
leaking those. There's definitely a nice way to avoid work here, but it
turns out it wasn't quite this trivial. Reverting for now.
Merge the load_elf() and commit_elf() functions into a single
load_main_executable() function that takes care of both things.
Also split "stage 3" into two separate stages, keeping the lazy
relocations in stage 3, and adding a stage 4 for calling library
initialization functions.
We also make sure to map the main executable before dealing with
any of its dependencies, to ensure that non-PIE executables get
loaded at their desired address.
Currently, when a process which has a tracee exits, nothing will happen,
leaving the tracee unable to be attached again. This will call the
stop_tracing function on any process which is traced by the exiting
process and sending the SIGSTOP signal making the traced process wait
for a SIGCONT (just as Linux does)
This patch combines inode the scan for an available inode with the
updating of the bit in the inode bitmap into a single operation.
We also exit the scan immediately when we find an inode, instead of
continuing until we've scanned all the eligible groups(!)
Finally, we stop holding the filesystem lock throughout the entire
operation, and instead only take it while actually necessary
(during inode allocation, flush, and inode cache update.)
Improve a bunch of situations where we'd previously panic the kernel
on failure. We now propagate whatever error we had instead. Usually
that'll be EIO.
Both inode and block allocation operate on bitmap blocks and update
counters in the superblock and group descriptor.
Since we're here, also add some error propagation around this code.
Refactors menubar creation to avoid a null parent window during
construction; moves search options to the more traditional edit
menu; creates and exclusive action group for bytes per row
Fixes#5177 in part
Commit 6a6f19a72 broke the cell position display in the top left of the
Spreadsheet window and the title of the cell type dialog, causing the
application to crash when interacting with cells beyond column FE.
This was another vestige from a long time ago, when exiting a thread
would mutate global data structures that were only protected by the
interrupt flag.
This was necessary in the past when crash handling would modify
various global things, but all that stuff is long gone so we can
simplify crashes by leaving the interrupt flag alone.
Rather than invalidating the entire window, which is very expensive on
the transparent desktop widget, just invalidate the areas that actually
need updating.
It's possible that pending invalidation rectangles haven't been
flushed when processing a paint event. Handle them right away,
which avoids another round trip.
Make more of the kernel compile in 64-bit mode, and make some things
pointer-size-agnostic (by using FlatPtr.)
There's a lot of work to do here before the kernel will even compile.
calloc() was internally calling malloc_impl() which would scrub out
all the allocated memory with the scrub byte (0xdc). We would then
immediately zero-fill the memory.
This was obviously a waste of time, and our hash tables were doing
it all the time. :^)
In line wrapping mode, the same column refers both to the end of a
visual line and the beginning of the subsequent visual line. This
meant that the cursor couldn't move to the beginning of a visual
line. Now, we only let the cursor go to the end of a physical line,
not the visual line.
Closes#4786
