RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-26 04:37:44 +00:00
Code Issues Activity Actions

Kernel/PCI: Don't hold spinlocks when doing fast device enumeration

Browse source 
Instead, hold the lock while we copy the contents to a stack-based
Vector then iterate on it without any locking.

Because we rely on heap allocations, we need to propagate errors back
in case of OOM condition, therefore, both PCI::enumerate API function
and PCI::Access::add_host_controller_and_enumerate_attached_devices use
now a ErrorOr<void> return value to propagate errors. OOM Error can only
occur when enumerating the m_device_identifiers vector under a spinlock
and trying to expand the temporary Vector which will be used locklessly
to actually iterate over the PCI::DeviceIdentifiers objects.
This commit is contained in:
Liav A 2022-02-04 19:48:13 +02:00 • committed by Andreas Kling
parent c0ed656c94
commit 3fb289e27d
14 changed files with 69 additions and 42 deletions
Download patch file Download diff file Expand all files Collapse all files

4
Kernel/Bus/PCI/API.h
View file

@ -6,6 +6,8 @@
#pragma once
#include <AK/Error.h>
#include <AK/Try.h>
#include <Kernel/Bus/PCI/Definitions.h>
namespace Kernel::PCI {
@ -19,7 +21,7 @@ u32 read32(Address address, PCI::RegisterOffset field);
HardwareID get_hardware_id(PCI::Address);
bool is_io_space_enabled(Address);
void enumerate(Function<void(DeviceIdentifier const&)> callback);
ErrorOr<void> enumerate(Function<void(DeviceIdentifier const&)> callback);
void enable_interrupt_line(Address);
void disable_interrupt_line(Address);
void raw_access(Address, u32, size_t, u32);