Refactor the kernel's mm module #145
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In doing so, we were overating over one page too many. Instead iterate over an exlusive range.
We have to assume PAGE_SIZE from the HAL to be a baked in constant at compile-time. Similar to a PAGE_SIZE define in a C project. Don't pass it because it won't (read *must not*) change at run-time.
n1tram1
force-pushed
the
refactor/kernel-mm-module
branch
from
5632e1a to
2835844
Compare
|
Hum, looks like I did not fix riscv alongside aarch64... |
Skallwar
reviewed
Oct 17, 2023
| @@ -97,12 +97,14 @@ fn test_timer_interrupt() -> TestResult { | |||
| fn test_pagetable_remap() -> TestResult { | |||
| info!("Testing the remapping capabilities of our pagetable..."); | |||
|
|
|||
| let page_src = alloc_pages(1).unwrap(); | |||
| let page_src = globals::PHYSICAL_MEMORY_MANAGER.alloc(1).unwrap(); | |||
| let page_src = unsafe { slice::from_raw_parts_mut(page_src as *mut u8, PAGE_SIZE) }; | |||
There was a problem hiding this comment.
Maybe we should add a wrapper for that because we do this casting in a lot of different places
| })?; | ||
| let pt = hal_core::mm::prefill_pagetable::<PageTable>(r, rw, rwx, pre_allocated, allocator)?; | ||
|
|
||
| // TODO: put into into the hal_core::Error |
There was a problem hiding this comment.
Do the todo now or later?
CohenArthur
reviewed
Nov 15, 2023
CohenArthur
approved these changes
Nov 15, 2023
Had to change a few interfaces to get around ownership issues. But now that all DRAM is not mapped as r/w, hopefully we will catch bugs earlier later on.
PageAllocFn was not very wieldy, the alloc_pages_* (alloc_pages_for_hal etc...) functions were a mess and it was hard to know what to use when etc... Use a trait instead, that way we can still pass an allocator that does not allocate in cases when we don't want to allocate, and pass the whole allocator when allocations actually are needed. That way it is easier to add functionality for when the HALs need more info from the allocator. Also it is much much easier to manage ownership&lifetimes this way.
Thanks to the previous patch, it was fairly easy to just make the PMM thread-safe. It also makes interfaces a bit nicer imo. Lastly I did it because it was easy and I felt like it.
n1tram1
force-pushed
the
refactor/kernel-mm-module
branch
from
2835844 to
e63a319
Compare
n1tram1
force-pushed
the
refactor/kernel-mm-module
branch
from
e63a319 to
7e73956
Compare
Skallwar
approved these changes
Nov 30, 2023
n1tram1
force-pushed
the
refactor/kernel-mm-module
branch
from
9b8b759 to
ce3d57e
Compare
n1tram1
force-pushed
the
refactor/kernel-mm-module
branch
from
ce3d57e to
72db5f4
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I was feeling like the
mmmodule was a hot mess ready to blow our legs off.Now that we only map the memory we need as r/w, I hope we will catch bugs earlier (if we create some).
I also prefer those interfaces, and having the PMM thread-safe is a nice step forward.
I should have cleaned up all the weird debug stuff I had left laying around.