diff --git a/scenario5-debug-symbols/README.md b/scenario5-debug-symbols/README.md
index fbb0659..d7fea91 100644
--- a/scenario5-debug-symbols/README.md
+++ b/scenario5-debug-symbols/README.md
@@ -72,22 +72,35 @@ With debug symbols, you can see exactly which lines are hot:
 
 ```bash
 perf record ./withdebug 500 5000
-perf annotate compute_inner
+perf annotate --stdio -l compute_inner
 ```
 
-This shows the source code with cycle counts per line!
+The `-l` flag shows a summary of hot source lines at the top:
 
 ```
-       │    double compute_inner(double x, int iterations) {
-       │        double result = x;
-  1.23 │        for (int i = 0; i < iterations; i++) {
- 45.67 │            result = sin(result) * cos(result) + sqrt(fabs(result));
- 23.45 │            result = result * 1.0001 + 0.0001;
-       │        }
-       │        return result;
-       │    }
+Sorted summary for file .../withdebug
+----------------------------------------------
+   72.75 program.c:29
+   27.25 program.c:28
 ```
 
+This tells you line 29 (`result = result * 1.0001 + 0.0001`) consumed ~73% of cycles!
+
+Below the summary, you see interleaved source and assembly:
+
+```
+ Percent | Source code & Disassembly of withdebug
+--------------------------------------------------
+         : 9    double compute_inner(double x, int iterations) {
+         : 10   double result = x;
+   27.25 :   130f:   addsd  %xmm0,%xmm1 // program.c:28
+   27.67 :   1313:   mulsd  ...         // program.c:29
+```
+
+**Note on line numbers**: The left margin (9, 10, etc.) shows objdump output
+line numbers, NOT source file lines. The actual source lines are shown in the
+summary and in `// program.c:XX` comments on assembly lines.
+
 ## Exercise 5: Understanding Symbol Tables
 
 ```bash