Mold: A Massively Parallel Linker

shameless plug: lld 23.x is significantly faster than 22.x! I wrote about the optimizations in https://maskray.me/blog/2026-04-12-recent-lld-elf-performanc... As a side note, parallel input file parsing and symbol resolution aren't a silver bullet. I have an experimental branch (net ~800 lines of code) adding this to lld/ELF, but it actually loses performance when restricted to --threads={1,2}

I read how mold implements parallel input file parsing a while ago. It is overly complicated. wild's model seems superior.

---

"Table 4." in the paper feels misleading to me. I have never measured such high speedups. In addition, on debug builds (especially --gdb-index builds), mold's edge over lld is smaller.

On ICF, some unknown fraction of mold's speedup is due to "doesn't do the symbol bootkeeping". The symbol tables are different in mold but identical in lld.

Here's a lwn.net link to Ian Lance Taylor's blog posts about his gold linker (his blog doesn't seem to have any table of contents or group the linker-related blog posts in one area/tag).

see https://lwn.net/Articles/276782/

GNU gold is deprecated and will probably be removed soon. ( https://www.phoronix.com/news/GNU-Gold-Linker-Deprecated )

But back in the late 2000s, the blog posts were a combination of algorithm optimization class and software history since no one seemed to step up to cleanup the technical debt that GNU ld had accumulated over the decades.

Old tweet from Rui, but I saw it in a recent zig video and really liked the sentiment:

(https://x.com/rui314/status/1560277403787997185)

> my strategy is to increase the value of commons by, say, 100 and get 1 as a return, so bear with me. If you want to use mold for free, you can just do that. This is my open-source business strategy.

Wishing the best for Mold and Rui

Wild is a decent bit faster than mold at this point:

https://github.com/wild-linker/wild

(Benchmarks in the README.)

This is a wonderful paper that includes many optimization tricks that are generally applicable, including various optimizations that they retrofitted into lld. But I thought there was one meta optimization that was especially interesting (from page 2):

> A new linker's early adopters knowingly accept such differences, which is one reason why drastic improvements tend to come from new linkers rather than existing ones

This is true of many things, not just linkers. I think this is encouraging for people to (re)try things that have already been done, because newcomers are allowed to take risks that established things can't take.

We use mold as the default linker in the Stagex linux distribution. It shaved hours off our full tree build time.

Would use this at a fairly large company if we had windows and Mac support, probably could get budget to pay for the features it’s missing.