diff --git a/compiler/rustc_data_structures/src/profiling.rs b/compiler/rustc_data_structures/src/profiling.rs index 9a85b9d02c995..f0b413c795e9c 100644 --- a/compiler/rustc_data_structures/src/profiling.rs +++ b/compiler/rustc_data_structures/src/profiling.rs @@ -590,24 +590,21 @@ pub fn print_time_passes_entry( end_rss: Option, ) { let rss_to_mb = |rss| (rss as f64 / 1_000_000.0).round() as usize; + let rss_change_to_mb = |rss| (rss as f64 / 1_000_000.0).round() as i128; let mem_string = match (start_rss, end_rss) { (Some(start_rss), Some(end_rss)) => { - // It's tempting to add the change in RSS from start to end, but its somewhat confusing - // and misleading when looking at time-passes output. Consider two adjacent entries: - // - // time: 10.000; rss start: 1000MB, end: 1000MB, change: 0MB pass1 - // time: 5.000; rss start: 2000MB, end: 2000MB, change: 0MB pass2 - // - // If you're looking for jumps in RSS based on the change column, you miss the fact - // that a 1GB jump happened between pass1 and pass2 (supposing pass1 and pass2 actually - // occur sequentially and pass1 isn't just nested within pass2). It's easy to imagine - // someone missing this or being confused by the fact that the change is zero. - - format!("; rss: {:>5}MB -> {:>5}MB", rss_to_mb(start_rss), rss_to_mb(end_rss)) + let change_rss = end_rss as i128 - start_rss as i128; + + format!( + "; rss: {:>4}MB -> {:>4}MB ({:>+5}MB)", + rss_to_mb(start_rss), + rss_to_mb(end_rss), + rss_change_to_mb(change_rss), + ) } - (Some(start_rss), None) => format!("; rss start: {:>5}MB", rss_to_mb(start_rss)), - (None, Some(end_rss)) => format!("; rss end: {:5>}MB", rss_to_mb(end_rss)), + (Some(start_rss), None) => format!("; rss start: {:>4}MB", rss_to_mb(start_rss)), + (None, Some(end_rss)) => format!("; rss end: {:>4}MB", rss_to_mb(end_rss)), (None, None) => String::new(), };