The get_dir_entries syscall failed if the serialized form of all the
directory entries together was too large to fit in its temporary buffer.
Now the kernel uses a fixed size buffer, that is flushed to an output
buffer when it is full. If this flushing operation fails because there
is not enough space available, the syscall will return -EINVAL. That
error code is then used in userspace as a signal to allocate a larger
buffer and retry the syscall.
When attempting to fix the dirent code I also changed
this to use strlcpy instead of the custom string copy
loop that was there before. Looking over strlcpy it
looked like it should work when using a non null terminated
string, I obviously misinterpreted the implementation
as it will read till it finds a null terminator.
Manually null terminate the string to address this.
Gunnar found this after he fixed UserspaceEmulator.
I reproduced it locally using his branch, and also
found the memory leak I had in the unit test for the
scandir that I added, so lets fix that as well.
Reported-by: Gunnar Beutner <gbeutner@serenityos.org>
I ran into a need for this when running stress-ng against the system.
This change implements the full functionality of scandir, where it
accepts a selection callback, as well as a comparison callback.
These can be used to trim and sort the entries from the directory
that we are being asked to enumerate. A test was also included to
validate the new functionality.
While adding new functionality which used the d_reclen member
to copy a dirent, I realized that the value being populated
was incorrect. sys_ent::total_size() function calculates the
size of the sys_ent structure, but dirent is larger than sys_ent.
This causes the malloc to be too small and you end up missing
the end of the copy, which can miss the null terminator
resulting in corrupt dirent names.
Since we don't actually use the variable length member nature
of dirent on other platforms we can just use the full size of
the struct ad the d_reclen value.
Also replace the custom strcpy with the standard version.
SPDX License Identifiers are a more compact / standardized
way of representing file license information.
See: https://spdx.dev/resources/use/#identifiers
This was done with the `ambr` search and replace tool.
ambr --no-parent-ignore --key-from-file --rep-from-file key.txt rep.txt *
(...and ASSERT_NOT_REACHED => VERIFY_NOT_REACHED)
Since all of these checks are done in release builds as well,
let's rename them to VERIFY to prevent confusion, as everyone is
used to assertions being compiled out in release.
We can introduce a new ASSERT macro that is specifically for debug
checks, but I'm doing this wholesale conversion first since we've
accumulated thousands of these already, and it's not immediately
obvious which ones are suitable for ASSERT.
This achieves two things:
- Programs can now intentionally perform arbitrary syscalls by calling
syscall(). This allows us to work on things like syscall fuzzing.
- It restricts the ability of userspace to make syscalls to a single
4KB page of code. In order to call the kernel directly, an attacker
must now locate this page and call through it.
