feat(riscv platform vf2): vf2 platform adapted to riscv architecture #1285

JensenWei007 · 2025-09-16T05:25:26Z

Add necessary platform driver support
Modify some startup processes and assert
Fixed some issues

fslongjin · 2025-09-25T08:34:54Z

bugbot run

cursor

This PR is being reviewed by Cursor Bugbot

Details

You are on the Bugbot Free tier. On this plan, Bugbot will review limited PRs each billing cycle.

To receive Bugbot reviews on all of your PRs, visit the Cursor dashboard to activate Pro and start your 14-day free trial.

Bug: Kernel Stack Initialization Mismatch

The init_idle function assumes processor 0 is the Boot Strap Processor (BSP) for kernel stack initialization. This conflicts with the removed assertion and the TODO comment noting that the boot hart might not be 0 on platforms like riscv-vf2. This mismatch can lead to incorrect kernel stack setup for the actual BSP and processor 0, potentially causing system instability.

kernel/src/process/idle.rs#L38-L46

DragonOS/kernel/src/process/idle.rs

Lines 38 to 46 in c1aa3d9

    
           for i in 0..PerCpu::MAX_CPU_NUM { 
        
               let kstack = if unlikely(i == 0) { 
        
                   let stack_ptr = 
        
                       VirtAddr::new(Self::stack_ptr().data() & (!(KernelStack::ALIGN - 1))); 
        
                   // 初始化bsp的idle进程 
        
                   let mut ks = unsafe { KernelStack::from_existed(stack_ptr) } 
        
                       .expect("Failed to create kernel stack struct for BSP."); 
        
                   unsafe { ks.clear_pcb(true) }; 
        
                   ks

cursor · 2025-09-25T08:38:26Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/mmc.rs

+        if cmd.data_expected() {
+            let buffer = // TODO: dirty
+                buffer.unwrap_or(unsafe { core::slice::from_raw_parts_mut(core::ptr::NonNull::dangling().as_ptr(), 64) });
+            assert!(buffer_offset == 0);


Bug: Invalid Memory Access in send_cmd Function

The send_cmd function creates a mutable slice from a dangling pointer when the buffer argument is None. This results in a 64-element slice pointing to invalid memory, which is then accessed during data transfers. This leads to undefined behavior, potential memory corruption, and system instability. This also bypasses the debug_assert on line 404 when data is expected.

cursor · 2025-09-25T08:38:26Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/mmc.rs

+        assert!(buf.len() == BLOCK_SIZE);
+
+        #[allow(mutable_transmutes)]
+        let buf = unsafe { core::mem::transmute(buf) };


Bug: Unsafe Transmute Violates Rust Memory Safety

The write_block function uses mem::transmute to convert an immutable &[u8] slice into a mutable &mut [usize] slice. This violates Rust's aliasing rules and memory safety, leading to undefined behavior. The transmute also changes the element type, which can cause issues with data interpretation and alignment.

cursor · 2025-09-25T08:38:26Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/registers.rs

+impl BUFADDRU {
+    pub fn offset() -> usize {
+        0xA0
+    }


Bug: Incorrect Offset for 64-bit Buffer Address Upper Part

The BUFADDRL and BUFADDRU structs incorrectly share the same offset (0xA0). As these represent the lower and upper 32-bit parts of a 64-bit buffer address, BUFADDRU needs a distinct offset, likely 0xA4, consistent with other 64-bit address register pairs.

cursor · 2025-09-25T08:38:26Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/mmc.rs

+        let base = self.virt_base_address() as *mut CTRL;
+        let ctrl = self.control_reg().with_fifo_reset(true);
+        unsafe { base.byte_add(*self.fifo_offset.get()).write_volatile(ctrl) }
+    }


Bug: FIFO Reset Error

In reset_fifo, the control register value is written to the FIFO data offset (0x600) instead of the correct control register offset (0x00). This prevents the FIFO from resetting and may corrupt other hardware registers.

Copilot

Pull Request Overview

This PR adds support for the VisionFive 2 (VF2) RISC-V platform by introducing platform-specific drivers and modifying startup procedures to accommodate hardware differences.

Key changes:

Implements VF2 SD/MMC card driver (DW MSHC controller)
Adjusts boot hart initialization logic for RISC-V platforms
Relocates architecture setup in the boot sequence for RISC-V

Reviewed Changes

Copilot reviewed 15 out of 16 changed files in this pull request and generated 8 comments.

Show a summary per file

File	Description
kernel/src/process/idle.rs	Temporarily disables boot hart assertion that assumes hart 0
kernel/src/mm/init.rs	Moves RISC-V `setup_arch()` call earlier in memory initialization
kernel/src/init/initial_kthread.rs	Adds VF2-specific init binary to search path
kernel/src/init/init.rs	Guards `setup_arch()` call to prevent duplicate execution on RISC-V
kernel/src/filesystem/vfs/vcore.rs	Adds `/dev/sdio1` to root filesystem device candidates
kernel/src/driver/base/device/device_number.rs	Adds MMC block device major number constant
kernel/src/arch/riscv64/pci/mod.rs	Disables PCI initialization for VF2 platform
kernel/src/arch/riscv64/driver/vf2/mod.rs	Implements VF2 MMC/SDIO probe and initialization
kernel/src/arch/riscv64/driver/vf2/dw_mshc/registers.rs	Defines register structures for DW MSHC controller
kernel/src/arch/riscv64/driver/vf2/dw_mshc/mod.rs	Module structure for DW MSHC driver components
kernel/src/arch/riscv64/driver/vf2/dw_mshc/mmc.rs	Core MMC driver implementation with block device interface
kernel/src/arch/riscv64/driver/vf2/dw_mshc/dma.rs	DMA descriptor structures for MSHC controller
kernel/src/arch/riscv64/driver/sbi.rs	Modifies console output to handle line endings for RustSBI
kernel/src/arch/riscv64/driver/mod.rs	Exports VF2 driver module
kernel/Cargo.toml	Adds byte-slice-cast dependency for driver

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

Copilot · 2025-10-22T05:37:32Z

kernel/src/process/idle.rs

+        // TODO: 这里需要修正一下判断逻辑
+        // 在 riscv-vf2 中, boot hart 不一定是0, 他可能是1, 因此会过不了


The TODO comment should be in English to maintain consistency with the codebase. Additionally, it should reference the specific issue or provide more context about when this will be addressed.

Suggested change

// TODO: 这里需要修正一下判断逻辑

// 在 riscv-vf2 中, boot hart 不一定是0, 他可能是1, 因此会过不了

// TODO: The logic here needs to be fixed. See issue #123 for details.

// In riscv-vf2, the boot hart is not always 0; it can be 1, so this check may fail.

Copilot · 2025-10-22T05:37:32Z

kernel/src/init/initial_kthread.rs

    ("/bin/busybox", Some("init")),
    ("/bin/init", None),
    ("/bin/sh", None),
+    ("/bin/riscv_rust_init", None), // 对vf2, 目前没做cmdline的适配, 加个默认寻找


The inline comment should be in English to match the project's language standards.

Suggested change

("/bin/riscv_rust_init", None), // 对vf2, 目前没做cmdline的适配, 加个默认寻找

("/bin/riscv_rust_init", None), // For vf2, cmdline adaptation is not yet implemented, so add a default search

Copilot · 2025-10-22T05:37:33Z

kernel/src/arch/riscv64/pci/mod.rs

-    let fdt = open_firmware_fdt_driver().fdt_ref()?;
+    // vf2 不需要, 事实上 qemu 也不使用 pci，设备都是使用 mmio
+    // 因此其实初始化这个没有太大的意义, 先注释掉
+    // TODO: 如果取消注释且启用 vf2 平台, 那么需要补充 vf2 的 pcie 驱动


These comments should be written in English to maintain consistency with the rest of the codebase.

Suggested change

// TODO: 如果取消注释且启用 vf2 平台, 那么需要补充 vf2 的 pcie 驱动

// TODO: If you uncomment this and enable the vf2 platform, you need to add the vf2 PCIe driver.

Copilot · 2025-10-22T05:37:33Z

kernel/src/arch/riscv64/driver/sbi.rs

-        for c in s {
-            #[allow(deprecated)]
-            sbi_rt::legacy::console_putchar(*c as usize);
+    // 原本的是适配opensbi的, 但是对rustsbi的适配存在问题, 不能正确的解析'\r'


This comment should be written in English to align with codebase documentation standards.

Suggested change

// 原本的是适配opensbi的, 但是对rustsbi的适配存在问题, 不能正确的解析'\r'

// The original implementation was adapted for OpenSBI, but there are compatibility issues with RustSBI, which cannot correctly parse '\r'

Copilot · 2025-10-22T05:37:33Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/mmc.rs

+    }
+
+    fn disk_range(&self) -> GeneralBlockRange {
+        // TODO: 实现自动读，下面的数字为fdisk -l DragonOS/bin/disk....img的结果


This TODO comment should be in English to maintain documentation consistency.

Suggested change

// TODO: 实现自动读，下面的数字为fdisk -l DragonOS/bin/disk....img的结果

// TODO: Implement automatic reading. The number below is the result of running 'fdisk -l DragonOS/bin/disk....img'.

Copilot · 2025-10-22T05:37:34Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/registers.rs

+    #[bits(16)]
+    pub oemid: usize, // OEM ID
+    #[bits(8)]
+    pub manfid: usize, // Manufacturer ID`


Corrected backtick at the end of the comment. Should be a regular comment without the trailing backtick.

Suggested change

pub manfid: usize, // Manufacturer ID`

pub manfid: usize, // Manufacturer ID

Copilot · 2025-10-22T05:37:34Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/mmc.rs

+    }
+
+    fn set_parent(&self, _parent: Weak<LockedDevFSInode>) {
+        panic!("DeviceINode for MMC is not supportted!")


Corrected spelling of 'supportted' to 'supported'.

Suggested change

panic!("DeviceINode for MMC is not supportted!")

panic!("DeviceINode for MMC is not supported!")

Copilot · 2025-10-22T05:37:34Z

kernel/src/arch/riscv64/driver/vf2/dw_mshc/dma.rs

+pub struct DES0 {
+    _reserved0: bool,
+    disable_interrupt_on_completion: bool,
+    last_decriptor: bool,


Corrected spelling of 'decriptor' to 'descriptor'.

Suggested change

last_decriptor: bool,

last_descriptor: bool,

- Add necessary platform driver support - Modify some startup processes and assert - Fixed some issues Signed-off-by: JensenWei007 <[email protected]>

Samuka007 · 2025-10-22T06:11:14Z

bugbot run

cursor · 2025-10-22T06:11:16Z

Skipping Bugbot: Unable to authenticate your request. Please make sure Bugbot is properly installed and configured for this repository.

Samuka007 · 2025-10-24T07:08:51Z

kernel/src/mm/init.rs

+
+    #[cfg(target_arch = "riscv64")]
+    crate::arch::init::setup_arch().expect("setup_arch failed");
+


这里似乎x86的初始化也可以迁到这个位置？不需要分开编译条件处理？ @fslongjin 怎么看
从原来的 kernel/src/init/init.rs:87 setup_arch()

JensenWei007 force-pushed the rv branch 2 times, most recently from 207fe36 to c1aa3d9 Compare September 16, 2025 05:45

cursor bot reviewed Sep 25, 2025

View reviewed changes

Copilot AI review requested due to automatic review settings October 22, 2025 05:36

JensenWei007 force-pushed the rv branch from c1aa3d9 to 8e2d23c Compare October 22, 2025 05:36

Copilot AI reviewed Oct 22, 2025

View reviewed changes

feat(riscv platform vf2): vf2 platform adapted to riscv architecture

952a100

- Add necessary platform driver support - Modify some startup processes and assert - Fixed some issues Signed-off-by: JensenWei007 <[email protected]>

JensenWei007 force-pushed the rv branch from 5849e33 to 952a100 Compare October 22, 2025 05:48

Samuka007 approved these changes Oct 24, 2025

View reviewed changes

Samuka007 reviewed Oct 24, 2025

View reviewed changes

	for i in 0..PerCpu::MAX_CPU_NUM {
	let kstack = if unlikely(i == 0) {
	let stack_ptr =
	VirtAddr::new(Self::stack_ptr().data() & (!(KernelStack::ALIGN - 1)));
	// 初始化bsp的idle进程
	let mut ks = unsafe { KernelStack::from_existed(stack_ptr) }
	.expect("Failed to create kernel stack struct for BSP.");
	unsafe { ks.clear_pcb(true) };
	ks

		// TODO: 这里需要修正一下判断逻辑
		// 在 riscv-vf2 中, boot hart 不一定是0, 他可能是1, 因此会过不了

	("/bin/riscv_rust_init", None), // 对vf2, 目前没做cmdline的适配, 加个默认寻找
	("/bin/riscv_rust_init", None), // For vf2, cmdline adaptation is not yet implemented, so add a default search

	// TODO: 如果取消注释且启用 vf2 平台, 那么需要补充 vf2 的 pcie 驱动
	// TODO: If you uncomment this and enable the vf2 platform, you need to add the vf2 PCIe driver.

	// 原本的是适配opensbi的, 但是对rustsbi的适配存在问题, 不能正确的解析'\r'
	// The original implementation was adapted for OpenSBI, but there are compatibility issues with RustSBI, which cannot correctly parse '\r'

	// TODO: 实现自动读，下面的数字为fdisk -l DragonOS/bin/disk....img的结果
	// TODO: Implement automatic reading. The number below is the result of running 'fdisk -l DragonOS/bin/disk....img'.

	pub manfid: usize, // Manufacturer ID`
	pub manfid: usize, // Manufacturer ID

	panic!("DeviceINode for MMC is not supportted!")
	panic!("DeviceINode for MMC is not supported!")


		#[cfg(target_arch = "riscv64")]
		crate::arch::init::setup_arch().expect("setup_arch failed");

Uh oh!

feat(riscv platform vf2): vf2 platform adapted to riscv architecture #1285

Are you sure you want to change the base?

feat(riscv platform vf2): vf2 platform adapted to riscv architecture #1285

Conversation

JensenWei007 commented Sep 16, 2025

Uh oh!

fslongjin commented Sep 25, 2025

Uh oh!

cursor bot left a comment

Choose a reason for hiding this comment

This PR is being reviewed by Cursor Bugbot

Bug: Kernel Stack Initialization Mismatch

Uh oh!

cursor bot Sep 25, 2025

Choose a reason for hiding this comment

Bug: Invalid Memory Access in send_cmd Function

Uh oh!

cursor bot Sep 25, 2025

Choose a reason for hiding this comment

Bug: Unsafe Transmute Violates Rust Memory Safety

Uh oh!

cursor bot Sep 25, 2025

Choose a reason for hiding this comment

Bug: Incorrect Offset for 64-bit Buffer Address Upper Part

Uh oh!

cursor bot Sep 25, 2025

Choose a reason for hiding this comment

Bug: FIFO Reset Error

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

Samuka007 commented Oct 22, 2025

Uh oh!

cursor bot commented Oct 22, 2025

Uh oh!

Samuka007 Oct 24, 2025

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Bug: Invalid Memory Access in `send_cmd` Function