RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-27 09:07:45 +00:00
Code Issues Activity Actions

Kernel: Move VirtIO code away from using a scatter gather list

Browse source 
Currently, when passing buffers into VirtIOQueues, we use scatter-gather
lists, which contain an internal vector of buffers. This vector is
allocated, filled and the destroy whenever we try to provide buffers
into a virtqueue, which would happen a lot in performance cricital code
(the main transport mechanism for certain paravirtualized devices).

This commit moves it over to using VirtIOQueueChains and building the
chain in place in the VirtIOQueue. Also included are a bunch of fixups
for the VirtIO Console device, making it use an internal VM::RingBuffer
instead.
This commit is contained in:
Sahan Fernando 2021-04-24 11:05:51 +10:00 • committed by Andreas Kling
parent 13d5cdcd08
commit ed0e7b53a5
11 changed files with 312 additions and 159 deletions
Download patch file Download diff file Expand all files Collapse all files

6
Kernel/VirtIO/VirtIOConsole.h
View file

@ -7,6 +7,7 @@
#pragma once
#include <Kernel/Devices/CharacterDevice.h>
#include <Kernel/VM/RingBuffer.h>
#include <Kernel/VirtIO/VirtIO.h>
namespace Kernel {
@ -25,6 +26,7 @@ public:
virtual ~VirtIOConsole() override;
private:
constexpr static size_t RINGBUFFER_SIZE = 2 * PAGE_SIZE;
virtual const char* class_name() const override { return m_class_name.characters(); }
virtual bool can_read(const FileDescription&, size_t) const override;
@ -38,8 +40,8 @@ private:
virtual String device_name() const override { return String::formatted("hvc{}", minor()); }
virtual void handle_queue_update(u16 queue_index) override;
OwnPtr<Region> m_receive_region;
OwnPtr<Region> m_transmit_region;
OwnPtr<RingBuffer> m_receive_buffer;
OwnPtr<RingBuffer> m_transmit_buffer;
static unsigned next_device_id;
};