aux: increase max payload size

src/libboard_artiq/src/drtioaux.rs

@@ -1,9 +1,10 @@
+use core::slice;
+
 use core_io::{Error as IoError, ErrorKind as IoErrorKind};
 use crc;
 use io::{proto::{ProtoRead, ProtoWrite},
          Cursor};
 use libboard_zynq::{time::Milliseconds, timer::GlobalTimer};
-use libcortex_a9::asm::dmb;
 
 pub use crate::drtioaux_proto::Packet;
 use crate::{drtioaux_proto::Error as ProtocolError, mem::mem::DRTIOAUX_MEM, pl::csr::DRTIOAUX};
@@ -56,19 +57,6 @@ pub fn has_rx_error(linkno: u8) -> bool {
     }
 }
 
-pub fn copy_work_buffer(src: *mut u32, dst: *mut u32, len: isize) {
-    // AXI writes must be 4-byte aligned (drtio proto doesn't care for that),
-    // and AXI burst reads/writes are not implemented yet in gateware
-    // thus the need for a work buffer for transmitting and copying it over
-    unsafe {
-        for i in 0..(len / 4) {
-            *dst.offset(i) = *src.offset(i);
-            //data memory barrier to prevent bursts
-            dmb();
-        }
-    }
-}
-
 fn receive<F, T>(linkno: u8, f: F) -> Result<Option<T>, Error>
 where F: FnOnce(&[u8]) -> Result<T, Error> {
     let linkidx = linkno as usize;
@@ -76,13 +64,7 @@ where F: FnOnce(&[u8]) -> Result<T, Error> {
         if (DRTIOAUX[linkidx].aux_rx_present_read)() == 1 {
             let read_ptr = (DRTIOAUX[linkidx].aux_read_pointer_read)() as usize;
             let ptr = (DRTIOAUX_MEM[linkidx].base + DRTIOAUX_MEM[linkidx].size / 2 + read_ptr * 0x400) as *mut u32;
-            // work buffer to accomodate axi burst reads
+            let result = f(slice::from_raw_parts(ptr as *mut u8, 0x400 as usize));
-            // buffer at maximum proto packet size, not maximum gateware supported size
-            // to minimize copying time
-            const LEN: usize = 512;
-            let mut buf: [u8; LEN] = [0; LEN];
-            copy_work_buffer(ptr, buf.as_mut_ptr() as *mut u32, LEN as isize);
-            let result = f(&buf);
             (DRTIOAUX[linkidx].aux_rx_present_write)(1);
             Ok(Some(result?))
         } else {
@@ -133,10 +115,7 @@ where F: FnOnce(&mut [u8]) -> Result<usize, Error> {
     unsafe {
         while (DRTIOAUX[linkno].aux_tx_read)() != 0 {}
         let ptr = DRTIOAUX_MEM[linkno].base as *mut u32;
-        // work buffer, works with unaligned mem access
+        let len = f(slice::from_raw_parts_mut(ptr as *mut u8, 0x400 as usize))?;
-        let mut buf: [u8; 1024] = [0; 1024];
-        let len = f(&mut buf)?;
-        copy_work_buffer(buf.as_mut_ptr() as *mut u32, ptr, len as isize);
         (DRTIOAUX[linkno].aux_tx_length_write)(len as u16);
         (DRTIOAUX[linkno].aux_tx_write)(1);
         Ok(())

src/libboard_artiq/src/drtioaux_async.rs

@@ -1,3 +1,5 @@
+use core::slice;
+
 use core_io::{Error as IoError, ErrorKind as IoErrorKind};
 use crc;
 use io::{proto::{ProtoRead, ProtoWrite},
@@ -8,7 +10,7 @@ use nb;
 use void::Void;
 
 pub use crate::drtioaux_proto::Packet;
-use crate::{drtioaux::{copy_work_buffer, has_rx_error, Error},
+use crate::{drtioaux::{has_rx_error, Error},
             mem::mem::DRTIOAUX_MEM,
             pl::csr::DRTIOAUX};
 
@@ -40,13 +42,7 @@ where F: FnOnce(&[u8]) -> Result<T, Error> {
         if (DRTIOAUX[linkidx].aux_rx_present_read)() == 1 {
             let read_ptr = (DRTIOAUX[linkidx].aux_read_pointer_read)() as usize;
             let ptr = (DRTIOAUX_MEM[linkidx].base + DRTIOAUX_MEM[linkidx].size / 2 + read_ptr * 0x400) as *mut u32;
-            // work buffer to accomodate axi burst reads
+            let result = f(slice::from_raw_parts(ptr as *mut u8, 0x400 as usize));
-            // buffer at maximum proto packet size, not maximum gateware supported size
-            // to minimize required copying time
-            const LEN: usize = 512;
-            let mut buf: [u8; LEN] = [0; LEN];
-            copy_work_buffer(ptr, buf.as_mut_ptr() as *mut u32, LEN as isize);
-            let result = f(&buf);
             (DRTIOAUX[linkidx].aux_rx_present_write)(1);
             Ok(Some(result?))
         } else {
@@ -106,10 +102,7 @@ where F: FnOnce(&mut [u8]) -> Result<usize, Error> {
     unsafe {
         let _ = block_async!(tx_ready(linkno)).await;
         let ptr = DRTIOAUX_MEM[linkno].base as *mut u32;
-        // work buffer, works with unaligned mem access
+        let len = f(slice::from_raw_parts_mut(ptr as *mut u8, 0x400 as usize))?;
-        let mut buf: [u8; 1024] = [0; 1024];
-        let len = f(&mut buf)?;
-        copy_work_buffer(buf.as_mut_ptr() as *mut u32, ptr, len as isize);
         (DRTIOAUX[linkno].aux_tx_length_write)(len as u16);
         (DRTIOAUX[linkno].aux_tx_write)(1);
         Ok(())

src/libboard_artiq/src/drtioaux_proto.rs

@@ -1,9 +1,11 @@
 use core_io::{Error as IoError, Read, Write};
 use io::proto::{ProtoRead, ProtoWrite};
 
+const MAX_PACKET: usize = 1024;
+
 // maximum size of arbitrary payloads
 // used by satellite -> master analyzer, subkernel exceptions
-pub const SAT_PAYLOAD_MAX_SIZE: usize  = /*max size*/512 - /*CRC*/4 - /*packet ID*/1 - /*last*/1 - /*length*/2;
+pub const SAT_PAYLOAD_MAX_SIZE: usize  = /*max size*/MAX_PACKET - /*CRC*/4 - /*packet ID*/1 - /*last*/1 - /*length*/2;
 // used by DDMA, subkernel program data (need to provide extra ID and destination)
 pub const MASTER_PAYLOAD_MAX_SIZE: usize = SAT_PAYLOAD_MAX_SIZE - /*source*/1 - /*destination*/1 - /*ID*/4;
 

src/libio/cursor.rs

@@ -45,7 +45,10 @@ impl<T: AsRef<[u8]>> Read for Cursor<T> {
     fn read(&mut self, buf: &mut [u8]) -> Result<usize, IoError> {
         let data = &self.inner.as_ref()[self.pos..];
         let len = buf.len().min(data.len());
-        buf[..len].copy_from_slice(&data[..len]);
+        // ``copy_from_slice`` generates AXI bursts, use a regular loop instead
+        for i in 0..len {
+            buf[i] = data[i];
+        }
         self.pos += len;
         Ok(len)
     }
@@ -55,7 +58,9 @@ impl Write for Cursor<&mut [u8]> {
     fn write(&mut self, buf: &[u8]) -> Result<usize, IoError> {
         let data = &mut self.inner[self.pos..];
         let len = buf.len().min(data.len());
-        data[..len].copy_from_slice(&buf[..len]);
+        for i in 0..len {
+            data[i] = buf[i];
+        }
         self.pos += len;
         Ok(len)
     }
@@ -68,7 +73,6 @@ impl Write for Cursor<&mut [u8]> {
 
 #[cfg(feature = "alloc")]
 impl Write for Cursor<Vec<u8>> {
-    #[inline]
     fn write(&mut self, buf: &[u8]) -> Result<usize, IoError> {
         self.inner.extend_from_slice(buf);
         Ok(buf.len())