nvme: better prp list

kernel/driver/block/nvme/src/queue.rs

@@ -93,28 +93,49 @@ pub struct QueuePair {
     inner: IrqSafeSpinlock<Inner>,
 }
 
-pub enum PrpList {
+pub struct PrpList {
-    None,
+    prp1: PhysicalRegionPage,
-    One(PhysicalAddress),
+    prp2: PhysicalRegionPage,
-    Two(PhysicalAddress, PhysicalAddress),
+    list: Option<PageBox<[PhysicalAddress]>>,
-    List(PhysicalAddress, PageBox<[PhysicalAddress]>),
 }
 
 impl PrpList {
+    pub const fn empty() -> Self {
+        Self {
+            prp1: PhysicalRegionPage::null(),
+            prp2: PhysicalRegionPage::null(),
+            list: None,
+        }
+    }
+
     pub fn from_buffer(base: PhysicalAddress, size: usize) -> Result<Self, NvmeError> {
         // TODO hardcoded page size
         if base.into_u64() % 0x1000 != 0 {
             todo!();
         }
+
         match size {
-            0 => Ok(Self::None),
+            0 => Ok(Self::empty()),
-            _ if size <= 0x1000 => Ok(Self::One(base)),
+            _ if size <= 0x1000 => Ok(Self {
-            _ if size <= 0x2000 => Ok(Self::Two(base, base.add(0x1000))),
+                prp1: PhysicalRegionPage::with_addr(base),
+                prp2: PhysicalRegionPage::null(),
+                list: None,
+            }),
+            _ if size <= 0x2000 => Ok(Self {
+                prp1: PhysicalRegionPage::with_addr(base),
+                prp2: PhysicalRegionPage::with_addr(base.add(0x1000)),
+                list: None,
+            }),
             _ => {
                 let count = (size + 0xFFF) / 0x1000;
                 let list = PageBox::new_slice_with(|i| base.add((i + 1) * 0x1000), count - 1)
                     .map_err(NvmeError::MemoryError)?;
-                Ok(Self::List(base, list))
+
+                Ok(Self {
+                    prp1: PhysicalRegionPage::with_addr(base),
+                    prp2: PhysicalRegionPage::with_addr(unsafe { list.as_physical_address() }),
+                    list: Some(list),
+                })
             }
         }
     }
@@ -339,25 +360,8 @@ impl QueuePair {
         let mut inner = self.inner.lock();
         let mut sqe = SubmissionQueueEntry::zeroed();
 
-        match ranges {
+        sqe.data_pointer[0] = ranges.prp1;
-            PrpList::None => {
+        sqe.data_pointer[1] = ranges.prp2;
-                sqe.data_pointer[0] = PhysicalRegionPage::null();
-                sqe.data_pointer[1] = PhysicalRegionPage::null();
-            }
-            &PrpList::One(buf0) => {
-                sqe.data_pointer[0] = PhysicalRegionPage::with_addr(buf0);
-                sqe.data_pointer[1] = PhysicalRegionPage::null();
-            }
-            &PrpList::Two(buf0, buf1) => {
-                sqe.data_pointer[0] = PhysicalRegionPage::with_addr(buf0);
-                sqe.data_pointer[1] = PhysicalRegionPage::with_addr(buf1);
-            }
-            PrpList::List(buf0, list) => {
-                sqe.data_pointer[0] = PhysicalRegionPage::with_addr(*buf0);
-                sqe.data_pointer[1] =
-                    PhysicalRegionPage::with_addr(unsafe { list.as_physical_address() });
-            }
-        }
 
         cmd.fill_sqe(&mut sqe);
 
@@ -374,7 +378,7 @@ impl QueuePair {
     }
 
     pub async fn request_no_data<C: Command>(&self, req: C) -> Result<(), NvmeError> {
-        let list = PrpList::None;
+        let list = PrpList::empty();
         let command_id = self.submit(req, &list, true)?;
         self.wait_for_completion(command_id, ())
             .await
@@ -399,8 +403,6 @@ impl QueuePair {
         let mut inner = self.inner.lock();
         let mut n = 0;
 
-        let mut completion_list = Vec::new();
-
         loop {
             let (cmp, expected_phase) = inner.cq.at_head(n);
             let cmp_phase = cmp.phase();
@@ -411,29 +413,24 @@ impl QueuePair {
 
             n += 1;
 
+            let command_id = cmp.command_id();
             let sub_queue_id = cmp.sub_queue_id();
             // TODO allow several sqs receive completions through one cq?
-            assert_eq!(sub_queue_id, self.id);
+            debug_assert_eq!(sub_queue_id, self.id);
 
             let sub_queue_head = cmp.sub_queue_head();
             let cmp = *cmp;
             inner.sq.take_until(sub_queue_head);
 
-            completion_list.push(cmp);
+            if inner.pending.remove(&command_id) {
+                inner.completed.insert(command_id, cmp);
+            }
         }
 
         if n != 0 {
             inner.cq.take(n);
         }
 
-        for cmp in completion_list {
-            let command_id = cmp.command_id();
-
-            if inner.pending.remove(&command_id) {
-                inner.completed.insert(command_id, cmp);
-            }
-        }
-
         if n != 0 {
             self.completion_notify.wake_all();
         }

kernel/driver/fs/ext2/src/file.rs

@@ -29,36 +29,11 @@ impl RegularNode {
         Ok(())
     }
 
-    async fn read(&self, mut pos: u64, buffer: &mut [u8]) -> Result<usize, Error> {
+    async fn read(&self, pos: u64, buffer: &mut [u8]) -> Result<usize, Error> {
         let holder = self.inode.get().await?;
         let inode = holder.read();
 
-        if pos >= inode.size(&self.fs) {
+        self.fs.read_inode_data(&inode, pos, buffer).await
-            return Ok(0);
-        }
-
-        let mut offset = 0;
-        let mut remaining = core::cmp::min(buffer.len(), (inode.size(&self.fs) - pos) as usize);
-
-        while remaining != 0 {
-            let block_index = pos / self.fs.block_size as u64;
-            let block_offset = (pos % self.fs.block_size as u64) as usize;
-            let amount = core::cmp::min(self.fs.block_size - block_offset, remaining);
-
-            self.fs
-                .with_inode_block(&inode, block_index as u32, |block| {
-                    buffer[offset..offset + amount]
-                        .copy_from_slice(&block[block_offset..block_offset + amount]);
-                    Ok(())
-                })
-                .await?;
-
-            pos += amount as u64;
-            offset += amount;
-            remaining -= amount;
-        }
-
-        Ok(offset)
     }
 
     async fn write(&self, mut pos: u64, buffer: &[u8]) -> Result<usize, Error> {

kernel/driver/fs/ext2/src/lib.rs

@@ -439,6 +439,41 @@ impl Ext2Fs {
         Ok(())
     }
 
+    pub async fn read_inode_data(
+        &self,
+        inode: &Inode,
+        mut pos: u64,
+        buffer: &mut [u8],
+    ) -> Result<usize, Error> {
+        let size = inode.size(self);
+
+        if pos >= size {
+            return Ok(0);
+        }
+
+        let mut offset = 0;
+        let mut remaining = buffer.len().min((size - pos) as usize);
+
+        while remaining != 0 {
+            let block_index = pos / self.block_size as u64;
+            let block_offset = (pos % self.block_size as u64) as usize;
+            let amount = remaining.min(self.block_size - block_offset);
+
+            self.with_inode_block(&inode, block_index as u32, |block| {
+                buffer[offset..offset + amount]
+                    .copy_from_slice(&block[block_offset..block_offset + amount]);
+                Ok(())
+            })
+            .await?;
+
+            pos += amount as u64;
+            offset += amount;
+            remaining -= amount;
+        }
+
+        Ok(offset)
+    }
+
     pub async fn flush_inode_cache(&self) -> Result<(), Error> {
         log::info!("Flushing inode cache");
         self.inode_cache.get().flush().await

kernel/driver/fs/ext2/src/symlink.rs

@@ -33,11 +33,12 @@ impl SymlinkNode {
         let inode = inode.read();
 
         let len = inode.size(&self.fs) as usize;
-        if len >= self.fs.block_size {
+        if len > self.fs.block_size {
-            todo!()
+            log::warn!("ext2: symlink size > block size");
+            return Err(Error::InvalidFile);
         }
         if buf.len() < len {
-            todo!();
+            return Err(Error::BufferTooSmall);
         }
 
         let mut write = self.cache.write();

kernel/libk/src/task/thread.rs

@@ -435,11 +435,13 @@ impl Thread {
     ///
     /// Will panic if no current thread is present. For try-style getter, see
     /// [Thread::get_current].
+    #[inline]
     pub fn current() -> CurrentThread {
         Self::get_current().unwrap()
     }
 
     /// Returns the current thread on the CPU, if any is present
+    #[inline]
     pub fn get_current() -> Option<CurrentThread> {
         // IrqGuard is held throughout
         let cpu = CpuImpl::<CpuQueue>::local();
@@ -449,6 +451,7 @@ impl Thread {
     }
 
     /// Returns a thread for given `id`, if such exists
+    #[inline]
     pub fn get(id: ThreadId) -> Option<Arc<Thread>> {
         THREADS.read().get(id).cloned()
     }

kernel/src/arch/x86_64/apic/local.rs

@@ -151,6 +151,7 @@ impl Device for LocalApic {
 }
 
 impl LocalApicInterface for LocalApic {
+    #[inline]
     fn clear_interrupt(&self) {
         self.regs.EndOfInterrupt.set(0);
     }

kernel/src/debug.rs

@@ -140,11 +140,15 @@ impl log::Log for RingLoggerSink {
 }
 
 impl log::Log for KernelLoggerSink {
-    fn enabled(&self, _metadata: &log::Metadata) -> bool {
+    fn enabled(&self, metadata: &log::Metadata) -> bool {
-        true
+        metadata.target() != "io"
     }
 
     fn log(&self, record: &log::Record) {
+        if !self.enabled(record.metadata()) {
+            return;
+        }
+
         if RING_AVAILABLE.load(Ordering::Acquire) {
             RING_LOGGER_SINK.log(record);
         }

userspace/sysutils/src/cat.rs

@@ -7,7 +7,7 @@ use std::{
 use sysutils::Input;
 
 fn cat_file(stdout: &mut Stdout, path: &str) -> io::Result<()> {
-    let mut buf = [0; 4096];
+    let mut buf = [0; 512];
     let mut reader = Input::open_str(path)?;
 
     loop {

userspace/sysutils/src/dd.rs

@@ -88,7 +88,7 @@ fn run<I: Read + Seek, O: Write>(
     let mut offset = 0;
     let mut bytes_read = 0;
 
-    // input.seek(SeekFrom::Start(src_position * src_block_size))?;
+    input.seek(SeekFrom::Start(src_position * src_block_size))?;
 
     let mut total_delta = Duration::ZERO;