drivers merge requestshttps://gitlab.redox-os.org/redox-os/drivers/-/merge_requests2024-03-28T13:57:08Zhttps://gitlab.redox-os.org/redox-os/drivers/-/merge_requests/151Use latest event queue interface everywhere.2024-03-28T13:57:08ZJacob Lorentzon4ldo2@protonmail.comUse latest event queue interface everywhere.https://gitlab.redox-os.org/redox-os/drivers/-/merge_requests/80Draft: Pcid scheme2022-05-27T21:55:33ZJacob Lorentzon4ldo2@protonmail.comDraft: Pcid schemeThis MR refactors `pcid` by separating it into two binaries: (1) `pcid`, which will enumerate the bus and then run as a traditional scheme-based driver daemon, and (2) pcid-spawner, which uses information from the `pci:` scheme to find m...This MR refactors `pcid` by separating it into two binaries: (1) `pcid`, which will enumerate the bus and then run as a traditional scheme-based driver daemon, and (2) pcid-spawner, which uses information from the `pci:` scheme to find matching drivers from a config file and spawn them. The pipe-based communication mechanism now uses a bidirectional pipe from the `pci:` scheme to replace the old `pipe2`-based one.
The motivation for this MR is to separate the driver logic from the subdriver spawning logic, and, a PCI scheme will probably also be useful for e.g. `lspci`, but mostly to have a better interface when acpid needs to be able to read from/write to PCI configuration space. In the future `pcid-spawner` might evolve into a more general driver manager, possibly more closely involved with init. If/when https://gitlab.redox-os.org/redox-os/kernel/-/merge_requests/194 is merged, the scheme can also provide IRQ handles directly from the kernel within a PCI device's "directory".https://gitlab.redox-os.org/redox-os/drivers/-/merge_requests/76Draft: Add a userspace RTC driver for x86 PCs2022-04-11T21:53:05ZJacob Lorentzon4ldo2@protonmail.comDraft: Add a userspace RTC driver for x86 PCsThe reason behind this is that an RTC driver should ideally consult ACPI
tables before assuming any I/O ports exist, which userspace can do much
better with the introduction of `acpid`.The reason behind this is that an RTC driver should ideally consult ACPI
tables before assuming any I/O ports exist, which userspace can do much
better with the introduction of `acpid`.