From 20d1729775d65074f6c22d654dec2a8aaa285f33 Mon Sep 17 00:00:00 2001 From: Rui Ueyama Date: Tue, 11 Jan 2022 00:12:12 +0900 Subject: [PATCH] Update README --- README.md | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index cb7fe3e220..620ebc2789 100644 --- a/README.md +++ b/README.md @@ -19,10 +19,10 @@ on a simulated 8-core 16-threads machine. | Firefox 89 libxul (1.64 GiB) | 32.95s | 6.80s | 1.42s mold is so fast that it is only 2x _slower_ than `cp` on the same -machine. +machine. Feel free to [file a bug](https://github.com/rui314/mold/issues) +if you find mold is not faster than other linkers. -Feel free to [file a bug](https://github.com/rui314/mold/issues) if -you find mold is not faster than other linkers. +mold currently supports x86-64, i386 and ARM64. ## Why does the speed of linking matter? @@ -189,7 +189,8 @@ If `mold` is in `.comment`, the file is created by mold. One reason is because it simply uses faster algorithms and efficient data structures than other linkers do. The other reason is that the -new linker is highly parallelized. +new linker is highly parallelized (i.e. optimized for _modern_ +multicore processors). Here is a side-by-side comparison of per-core CPU usage of lld (left) and mold (right). They are linking the same program, Chromium