[libc] Factor specifics of packet type out of process

[JonChesterfield]

Summary

NFC. Simplifies process slightly, gives more options for testing it.

[JonChesterfield]

libc/src/__support/RPC/rpc.h

255

diff hasn't gone well here - moving the methods into free functions changes where they are in the file. The indexing functions changed insofar as Packet<lane_size> is now Packet

[JonChesterfield]

• avoid moving functions in file to decrease diff size

[JonChesterfield]

• now compiles, moved invoke back to free functions

[jhuber6]

LG, thanks

libc/src/__support/RPC/rpc.h

253

LLVM style, static over anonymous namespace. ditto below.

[JonChesterfield]

• namespace => static

libc/src/__support/RPC/rpc.h

 /// otherwise it is owned by the server.
 ///
 /// This process is designed to allow the client and the server to exchange data
 /// using a fixed size packet in a mostly arbitrary order using the 'send' and
 /// 'recv' operations. The following restrictions to this scheme apply:
 ///   - The client will always start with a 'send' operation.
 ///   - The server will always start with a 'recv' operation.
 ///   - Every 'send' or 'recv' call is mirrored by the other process.
-template <bool Invert, uint32_t lane_size> struct Process {
+template <bool Invert, template Packet> struct Process {
   LIBC_INLINE Process() = default;
   LIBC_INLINE Process(const Process &) = delete;
   LIBC_INLINE Process &operator=(const Process &) = delete;
   LIBC_INLINE Process(Process &&) = default;
   LIBC_INLINE Process &operator=(Process &&) = default;
   LIBC_INLINE ~Process() = default;
 
   uint64_t port_count;
   cpp::Atomic<uint32_t> *inbox;
   cpp::Atomic<uint32_t> *outbox;
-  Packet<lane_size> *packet;
+  Packet *packet;
 
   cpp::Atomic<uint32_t> lock[DEFAULT_PORT_COUNT] = {0};
 
   /// Initialize the communication channels.
   LIBC_INLINE void reset(uint64_t port_count, void *buffer) {
     this->port_count = port_count;
     this->inbox = reinterpret_cast<cpp::Atomic<uint32_t> *>(
         advance(buffer, inbox_offset(port_count)));
     this->outbox = reinterpret_cast<cpp::Atomic<uint32_t> *>(
         advance(buffer, outbox_offset(port_count)));
-    this->packet = reinterpret_cast<Packet<lane_size> *>(
-        advance(buffer, buffer_offset(port_count)));
+    this->packet =
+        reinterpret_cast<Packet *>(advance(buffer, buffer_offset(port_count)));
   }
 
   /// Returns the beginning of the unified buffer. Intended for initializing the
   /// client after the server has been started.
   LIBC_INLINE void *get_buffer_start() const { return Invert ? outbox : inbox; }
 
   /// Allocate a memory buffer sufficient to store the following equivalent
   /// representation in memory.
[[clang::convergent]] LIBC_INLINE void unlock(uint64_t lane_mask,
     // Must restrict to a single thread to avoid one thread dropping the lock,
     // then an unrelated warp claiming the lock, then a second thread in this
     // warp dropping the lock again.
     uint32_t and_mask = ~(rpc::is_first_lane(lane_mask) ? 1 : 0);
     lock[index].fetch_and(and_mask, cpp::MemoryOrder::RELAXED);
     gpu::sync_lane(lane_mask);
   }
 
-  /// Invokes a function accross every active buffer across the total lane size.
-  LIBC_INLINE void invoke_rpc(cpp::function<void(Buffer *)> fn,
-                              Packet<lane_size> &packet) {
-    if constexpr (is_process_gpu()) {
-      fn(&packet.payload.slot[gpu::get_lane_id()]);
-    } else {
-      for (uint32_t i = 0; i < lane_size; i += gpu::get_lane_size())
-        if (packet.header.mask & 1ul << i)
-          fn(&packet.payload.slot[i]);
-    }
-  }
+  /// Number of bytes to allocate for an inbox or outbox.
+  LIBC_INLINE static constexpr uint64_t mailbox_bytes(uint64_t port_count) {
+    return port_count * sizeof(cpp::Atomic<uint32_t>);
+  }
+
+  /// Number of bytes to allocate for the buffer containing the packets.
+  LIBC_INLINE static constexpr uint64_t buffer_bytes(uint64_t port_count) {
+    return port_count * sizeof(Packet);
+  }
+
+  /// Offset of the inbox in memory. This is the same as the outbox if inverted.
+  LIBC_INLINE static constexpr uint64_t inbox_offset(uint64_t port_count) {
+    return Invert ? mailbox_bytes(port_count) : 0;
+  }
+
+  /// Offset of the outbox in memory. This is the same as the inbox if inverted.
+  LIBC_INLINE static constexpr uint64_t outbox_offset(uint64_t port_count) {
+    return Invert ? 0 : mailbox_bytes(port_count);
+  }
+
+  /// Offset of the buffer containing the packets after the inbox and outbox.
+  LIBC_INLINE static constexpr uint64_t buffer_offset(uint64_t port_count) {
+    return align_up(2 * mailbox_bytes(port_count), alignof(Packet));
+  }
+};
+
+namespace {
+
+template <uint32_t lane_size>
 
-  /// Alternate version that also provides the index of the current lane.
-  LIBC_INLINE void invoke_rpc(cpp::function<void(Buffer *, uint32_t)> fn,
-                              Packet<lane_size> &packet) {
-    if constexpr (is_process_gpu()) {
-      fn(&packet.payload.slot[gpu::get_lane_id()], gpu::get_lane_id());
-    } else {
-      for (uint32_t i = 0; i < lane_size; i += gpu::get_lane_size())
-        if (packet.header.mask & 1ul << i)
-          fn(&packet.payload.slot[i], i);
-    }
-  }
+template <uint32_t lane_size>
 
-  /// Number of bytes to allocate for an inbox or outbox.
-  LIBC_INLINE static constexpr uint64_t mailbox_bytes(uint64_t port_count) {
-    return port_count * sizeof(cpp::Atomic<uint32_t>);
-  }
-
-  /// Number of bytes to allocate for the buffer containing the packets.
-  LIBC_INLINE static constexpr uint64_t buffer_bytes(uint64_t port_count) {
-    return port_count * sizeof(Packet<lane_size>);
-  }
-
-  /// Offset of the inbox in memory. This is the same as the outbox if inverted.
-  LIBC_INLINE static constexpr uint64_t inbox_offset(uint64_t port_count) {
-    return Invert ? mailbox_bytes(port_count) : 0;
-  }
-
-  /// Offset of the outbox in memory. This is the same as the inbox if inverted.
-  LIBC_INLINE static constexpr uint64_t outbox_offset(uint64_t port_count) {
-    return Invert ? 0 : mailbox_bytes(port_count);
-  }
-
-  /// Offset of the buffer containing the packets after the inbox and outbox.
-  LIBC_INLINE static constexpr uint64_t buffer_offset(uint64_t port_count) {
-    return align_up(2 * mailbox_bytes(port_count), alignof(Packet<lane_size>));
-  }
-};
+} // namespace
 
 /// The port provides the interface to communicate between the multiple
 /// processes. A port is conceptually an index into the memory provided by the
 /// underlying process that is guarded by a lock bit.
 template <bool T, uint32_t S> struct Port {
   LIBC_INLINE Port(Process<T, S> &process, uint64_t lane_mask, uint64_t index,
                    uint32_t out)
       : process(process), lane_mask(lane_mask), index(index), out(out),
struct Client {
   template <uint16_t opcode> LIBC_INLINE cpp::optional<Port> try_open();
   template <uint16_t opcode> LIBC_INLINE Port open();
 
   LIBC_INLINE void reset(uint64_t port_count, void *buffer) {
     process.reset(port_count, buffer);
   }
 
 private:
-  Process<false, gpu::LANE_SIZE> process;
+  Process<false, Packet<gpu::LANE_SIZE>> process;
 };
 
 /// The RPC server used to respond to the client.
 template <uint32_t lane_size> struct Server {
   LIBC_INLINE Server() = default;
   LIBC_INLINE Server(const Server &) = delete;
   LIBC_INLINE Server &operator=(const Server &) = delete;
   LIBC_INLINE ~Server() = default;
 
   using Port = rpc::Port<true, lane_size>;
   LIBC_INLINE cpp::optional<Port> try_open();
   LIBC_INLINE Port open();
 
   LIBC_INLINE void reset(uint64_t port_count, void *buffer) {
     process.reset(port_count, buffer);
   }
 
   LIBC_INLINE void *get_buffer_start() const {
     return process.get_buffer_start();
   }
 
   LIBC_INLINE static uint64_t allocation_size(uint64_t port_count) {
-    return Process<true, lane_size>::allocation_size(port_count);
+    return Process<true, Packet<lane_size>>::allocation_size(port_count);
   }
 
 private:
-  Process<true, lane_size> process;
+  Process<true, Packet<lane_size>> process;
 };
 
 /// Applies \p fill to the shared buffer and initiates a send operation.
 template <bool T, uint32_t S>
 template <typename F>
 LIBC_INLINE void Port<T, S>::send(F fill) {
   uint32_t in = owns_buffer ? out ^ T : process.load_inbox(index);