We have many places in the kernel code that we have boolean flags that
are only set once, and never reset again but are checked multiple times
before and after the time they're being set, which matches the purpose
of the SetOnce class.
There are now 2 separate classes for almost the same object type:
- EnumerableDeviceIdentifier, which is used in the enumeration code for
all PCI host controller classes. This is allowed to be moved and
copied, as it doesn't support ref-counting.
- DeviceIdentifier, which inherits from EnumerableDeviceIdentifier. This
class uses ref-counting, and is not allowed to be copied. It has a
spinlock member in its structure to allow safely executing complicated
IO sequences on a PCI device and its space configuration.
There's a static method that allows a quick conversion from
EnumerableDeviceIdentifier to DeviceIdentifier while creating a
NonnullRefPtr out of it.
The reason for doing this is for the sake of integrity and reliablity of
the system in 2 places:
- Ensure that "complicated" tasks that rely on manipulating PCI device
registers are done in a safe manner. For example, determining a PCI
BAR space size requires multiple read and writes to the same register,
and if another CPU tries to do something else with our selected
register, then the result will be a catastrophe.
- Allow the PCI API to have a united form around a shared object which
actually holds much more data than the PCI::Address structure. This is
fundamental if we want to do certain types of optimizations, and be
able to support more features of the PCI bus in the foreseeable
future.
This patch already has several implications:
- All PCI::Device(s) hold a reference to a DeviceIdentifier structure
being given originally from the PCI::Access singleton. This means that
all instances of DeviceIdentifier structures are located in one place,
and all references are pointing to that location. This ensures that
locking the operation spinlock will take effect in all the appropriate
places.
- We no longer support adding PCI host controllers and then immediately
allow for enumerating it with a lambda function. It was found that
this method is extremely broken and too much complicated to work
reliably with the new paradigm being introduced in this patch. This
means that for Volume Management Devices (Intel VMD devices), we
simply first enumerate the PCI bus for such devices in the storage
code, and if we find a device, we attach it in the PCI::Access method
which will scan for devices behind that bridge and will add new
DeviceIdentifier(s) objects to its internal Vector. Afterwards, we
just continue as usual with scanning for actual storage controllers,
so we will find a corresponding NVMe controllers if there were any
behind that VMD bridge.
Instead of using a clunky if-statement paradigm, we now have all drivers
being declaring two methods for their adapter class - create and probe.
These methods are linked in each PCINetworkDriverInitializer structure,
in a new s_initializers static list of them.
Then, when we probe for a PCI device, we use each probe method and if
there's a match, then the corresponding create method is called. After
the adapter instance is created, we call the virtual initialize method
on it, because many drivers actually require a sort of post-construction
initialization sequence to ensure the network adapter can properly
function.
As a result of this change, it's much more easy to add more drivers and
the initialization code is more readable and it's easier to understand
when and where things could fail in the whole initialization sequence.
Instead of allocating after the construction point ensure that all Intel
drivers are allocating necessary buffer regions and then pass them to
the constructors.
This could let us fail early in case of OOM, so we don't touch a network
adapter before we ensure we have all the appropriate mappings in place.
A virtual method named device_name() was added to
Kernel::PCI to support logging the PCI::Device name
and address using dmesgln_pci. Previously, PCI::Device
did not store the device name.
All devices inheriting from PCI::Device now use dmesgln_pci where
they previously used dmesgln.
We now move the ErrorOr returning functions in the constructor to the
try_to_initialize() factory, which allows us to handle the errors and
removes two FIXME's :))
When scanning for network adapters, we give each driver a chance to
claim the PCI device and whoever claims it first gets to keep it.
Before this patch, the driver API returned a LockRefPtr<AdapterType>,
which made it impossible to propagate errors that occurred during
detection and/or initialization.
This patch changes the API so that errors can bubble all the way out
the PCI enumeration in NetworkingManagement::initialize() where we
perform all the network adapter auto-detection on boot.
When we eventually start to support hot-plugging network adapter in the
future, it will be even more important to propagate errors instead of
swallowing them.
Importantly, before this patch, some errors were "handled" by panicking
the kernel. This is no longer the case.
7 FIXMEs were killed in the making of this commit. :^)
This class is intended to replace all IOAddress usages in the Kernel
codebase altogether. The idea is to ensure IO can be done in
arch-specific manner that is determined mostly in compile-time, but to
still be able to use most of the Kernel code in non-x86 builds. Specific
devices that rely on x86-specific IO instructions are already placed in
the Arch/x86 directory and are omitted for non-x86 builds.
The reason this works so well is the fact that x86 IO space acts in a
similar fashion to the traditional memory space being available in most
CPU architectures - the x86 IO space is essentially just an array of
bytes like the physical memory address space, but requires x86 IO
instructions to load and store data. Therefore, many devices allow host
software to interact with the hardware registers in both ways, with a
noticeable trend even in the modern x86 hardware to move away from the
old x86 IO space to exclusively using memory-mapped IO.
Therefore, the IOWindow class encapsulates both methods for x86 builds.
The idea is to allow PCI devices to be used in either way in x86 builds,
so when trying to map an IOWindow on a PCI BAR, the Kernel will try to
find the proper method being declared with the PCI BAR flags.
For old PCI hardware on non-x86 builds this might turn into a problem as
we can't use port mapped IO, so the Kernel will gracefully fail with
ENOTSUP error code if that's the case, as there's really nothing we can
do within such case.
For general IO, the read{8,16,32} and write{8,16,32} methods are
available as a convenient API for other places in the Kernel. There are
simply no direct 64-bit IO API methods yet, as it's not needed right now
and is not considered to be Arch-agnostic too - the x86 IO space doesn't
support generating 64 bit cycle on IO bus and instead requires two 2
32-bit accesses. If for whatever reason it appears to be necessary to do
IO in such manner, it could probably be added with some neat tricks to
do so. It is recommended to use Memory::TypedMapping struct if direct 64
bit IO is actually needed.
Until now, our kernel has reimplemented a number of AK classes to
provide automatic internal locking:
- RefPtr
- NonnullRefPtr
- WeakPtr
- Weakable
This patch renames the Kernel classes so that they can coexist with
the original AK classes:
- RefPtr => LockRefPtr
- NonnullRefPtr => NonnullLockRefPtr
- WeakPtr => LockWeakPtr
- Weakable => LockWeakable
The goal here is to eventually get rid of the Lock* classes in favor of
using external locking.
Each of these strings would previously rely on StringView's char const*
constructor overload, which would call __builtin_strlen on the string.
Since we now have operator ""sv, we can replace these with much simpler
versions. This opens the door to being able to remove
StringView(char const*).
No functional changes.
We were frequently dropping packets when downloading large files.
Then we had to wait for TCP retransmission which slowed things down.
This patch dramatically improves E1000 throughput by increasing the
number of RX/TX buffers from 32/8 to 256/256.
The largest chunk of JavaScript from Discord now downloads in roughly
1 second instead of 7 seconds. :^)
Calls to link_up() in the E1000 driver would read the link state
directly from the hardware on every call. This had negative
performance impact in high throughput situations since link_up()
is called every time an IP packet's route is resolved.
This patch takes inspiration from the RTL8139 network adapter where
the link state is stored in a bool and only updated when the hardware
generates an interrupt related to link state change.
After this change I measured a ~9% increase in TCP Tx throughput
using:
cat /dev/zero | nc <host_IP> <host_port> from the Serenity VM to my
host machine