-
-
Notifications
You must be signed in to change notification settings - Fork 182
Fixes and improvements in GIC driver #910
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fixes and improvements in GIC driver #910
Conversation
kevinaboos
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Everything looks generally fine, I just made a few suggestions for minor improvements.
kevinaboos
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks solid, just left one comment clarifying SpiDestination behavior
* New `arm_boards` crate: per-board definitions for aarch64 builds
* The default is for QEMU's basic `virt` machine spec.
* Currently, this specifies the number of CPUs, their IDs, and
the interrupt controller configuration including one GIC
redistributor base address per core.
* The `gic` crate now uses this instead of defining its own
internal configuration values specific to QEMU.
* `multicore_bringup` now uses the list of `CpuId`s from
the selected ARM board configuration.
* The redistributor initialization routine code now enables
the dispatch of "1 of N"-distributed SPIs.
* All CPUs' redistributors are now initialized by the
bootstrap processor as part of `interrupts::init()`.
* Incorporates two fixes from #910:
* Fix a bug in `get_spi_target` & `set_spi_target` where
an atomic u32 read/write was used to manipulate a
u64 MMIO register.
* Introduce `Offset32` and `Offset64` types to distinguish
32-bit and 64-bit registers more clearly.
Co-authored-by: Kevin Boos <kevinaboos@gmail.com>
* New `arm_boards` crate: per-board definitions for aarch64 builds
* The default is for QEMU's basic `virt` machine spec.
* Currently, this specifies the number of CPUs, their IDs, and
the interrupt controller configuration including one GIC
redistributor base address per core.
* The `gic` crate now uses this instead of defining its own
internal configuration values specific to QEMU.
* `multicore_bringup` now uses the list of `CpuId`s from
the selected ARM board configuration.
* The redistributor initialization routine code now enables
the dispatch of "1 of N"-distributed SPIs.
* All CPUs' redistributors are now initialized by the
bootstrap processor as part of `interrupts::init()`.
* Incorporates two fixes from #910:
* Fix a bug in `get_spi_target` & `set_spi_target` where
an atomic u32 read/write was used to manipulate a
u64 MMIO register.
* Introduce `Offset32` and `Offset64` types to distinguish
32-bit and 64-bit registers more clearly.
Co-authored-by: Kevin Boos <kevinaboos@gmail.com> d2ed7ac
88ab729 to
25e20a7
Compare
kevinaboos
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
left a few comments about iterator usage and type conversion implementations
Small changes to the
giccrate:SpiDestination&IpiTargetCpu, which were previouslyTargetCpuTryFrom<u64>forMpidrValueto identify existing CPUs based on GIC register contents