fixup! Store Tachyon samples in bounded columnar blocks

Update the replay progress bar every 1,000 decoded samples instead of
only once per block. This matches the previous behavior and prevents the
bar from appearing stuck while large blocks are replayed.

Clarify in NEWS that Python 3.16 cannot read profiles written in the
previous binary format.

Author: ivonastojanovic

Lib/test/test_profiling/test_sampling_profiler/test_binary_format.py

@@ -1175,26 +1175,20 @@ def test_replay_byte_swapped_file(self):
             self.assertEqual(reader.replay_samples(collector), 6)
         self.assert_samples_equal(samples, collector)
 
-    def test_progress_follows_collected_samples(self):
-        """Progress never counts samples the collector has not received."""
+    def test_progress_updates_across_blocks(self):
+        """Progress is reported at fixed intervals across block boundaries."""
         samples, filename = self._write_block_spanning_file()
-
-        class CountingCollector:
-            received = 0
-
-            def collect(self, stack_frames, timestamps_us):
-                self.received += len(timestamps_us)
-
-        collector = CountingCollector()
         calls = []
         with BinaryReader(filename) as reader:
             reader.replay_samples(
-                collector,
-                lambda current, total: calls.append(
-                    (current, collector.received)))
-        self.assertGreater(len(calls), 2)
-        for current, received in calls:
-            self.assertEqual(current, received)
+                RawCollector(),
+                lambda current, total: calls.append((current, total)))
+        total = len(samples)
+        self.assertEqual(
+            calls,
+            [(0, total), (1000, total), (2000, total), (3000, total),
+             (4000, total), (total, total)],
+        )
 
     def test_reader_replays_twice(self):
         """Replaying twice gives the same samples and statistics."""

Misc/NEWS.d/next/Library/2026-10-20-12-49-32.gh-issue-159207.Lccezg.rst

@@ -1,2 +1,3 @@
 The :mod:`profiling.sampling` binary output format now stores samples in
-independent blocks.
+independent blocks. This is a breaking format change: Python 3.16 does not
+support reading profiles written in the previous format.

Modules/_remote_debugging/binary_io_reader.c

@@ -1498,7 +1498,6 @@ replay_block(BinaryReader *reader, RemoteDebuggingState *state,
             reader->sample_count - (uint64_t)*replayed) {
         return block_error("more samples than the file header declares");
     }
-    Py_ssize_t first_sample = *replayed;
     int result = 0;
     for (uint32_t sample = 0; sample < block->header.sample_count; sample++) {
         if (replay_sample(reader, state, collector, block, sample) < 0) {
@@ -1507,6 +1506,12 @@ replay_block(BinaryReader *reader, RemoteDebuggingState *state,
         }
         (*replayed)++;
         reader->stats.total_samples++;
+        if (*replayed % PROGRESS_CALLBACK_INTERVAL == 0 &&
+            invoke_progress_callback(progress_callback, *replayed,
+                                     reader->sample_count) < 0) {
+            result = -1;
+            break;
+        }
     }
     if (result == 0) {
         result = validate_block_consumed(reader, block);
@@ -1522,14 +1527,6 @@ replay_block(BinaryReader *reader, RemoteDebuggingState *state,
         Py_CLEAR(stream->batch);
     }
 
-    /* Report progress once the block's samples reached the collector. */
-    if (result == 0 &&
-        *replayed / PROGRESS_CALLBACK_INTERVAL !=
-            first_sample / PROGRESS_CALLBACK_INTERVAL &&
-        invoke_progress_callback(progress_callback, *replayed,
-                                 reader->sample_count) < 0) {
-        result = -1;
-    }
     return result;
 }