mold 2.0.0

Mold 2.0.0 is a new major release of our high-speed linker. With this release, we've transitioned our license from AGPL to MIT, aiming to expand the user base of our linker. This was not an easy decision, as those who have been following our progress know that we've been attempting to monetize our product through an AGPL/commercial license dual-licensing scheme. Unfortunately, this approach didn't meet our expectations. The license change represents our acceptance of this reality. We don't want to persist with a strategy that didn't work well.

As always, we welcome new GitHub sponsors. If you are happy with the license change, please consider becoming a sponsor.

In addition to the license change, here is a list of updates we have made in this release:

• Previously, mold could not produce an object file with more than 65520 sections using the --relocatable option. Now the bug has been fixed. (2e8bd0b)
• mold now interprets -undefined as a synonym for --undefined instead of -u ndefined. This seems inconsistent, as -ufoo is generally treated as -u foo (which is an alias for --undefined foo), but this is the behavior of the GNU linkers and LLVM lld, so we prioritize compatibility over consistency.
• -nopie is now handled as a synonym for --no-pie.
• [RISC-V] R_RISCV_SET_ULEB128 and R_RISCV_SUB_ULEB128 relocation types are now supported (4bffe26, 1ac5fe7)
• [PPC64] R_PPC64_REL32 relocation type is now supported. (ebd780e)

mold 1.11.0

mold 1.11.0 is a new release of the high-speed linker. This is not a big release but includes general improvements and bug fixes.

New features

• IBM Power10 has been supported. Previously, mold created broken executables for that target. (5065547)
• --hash-style=none has been added to cancel --hash-style=sysv, --hash-style=gnu or --hash-style=both. (ec75633)
• [ARM32] R_ARM_PLT32 relocation type has been supported. (e505900)
• [RISC-V] R_RISCV_PLT32 relocation type has been supported. (51845ac)

Bug fixes and compatibility improvements

• Previous versions of mold failed to link some programs in rare corner cases if Link-Time Optimization (LTO) is enabled. These bugs have been fixed. (e1a7590, 62d6537)
• mold used to ignore dependencies between DSOs. Since this version, if a required DSO depends on other as-needed DSO, mold keeps the latter DSO as a required one. This improves compatibility with GNU linkers. (1adde7a)
• [x86-64] mold can now link object files generated by old buggy versions of GCC. (d2970e0)
• [x86-64] Previously, a program with a very large .bss section may fail to link due to R_X86_64_REX_GOTPCRELX relocation overflow (#975). This bug has been fixed. (627bf7c)

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• Wei Wu
• Cybozu, Inc.
• Signal Slot Inc.
• Mercury
• 300baud
• G-Research-OSS
• kyle-elliott
• Jinkyu Yi
• Sanjeev
• Johan Andersson
• nasa
• Luke Kasz
• Matt Godbolt
• Brett Slatkin
• Dougall Johnson
• Santiago Pastorino
• JoJo
• daquexian
• Markus Schirp
• Javier Blazquez
• Jonathan Goodman
• Kristof Mulier
• Sebastian Dröge

mold 1.10.1

mold 1.10.1 contains only the following bug fix:

• mold 1.10.0 had a buffer overrun bug that causes the linker to terminate immediately if compiled with -D_GLIBCXX_ASSERTIONS. We fixed the unsafe memory access in this release. (7e65546)

mold 1.10.0

New features

• mold now officially supports the --print-dependencies option to print out dependency information between input files. Here is a truncated example output when linking mold itself with the option. There are many use cases of the option; for example, if you want to eliminate the dependency to some library from your program, you can use this option to find out all the functions that use the library's function to fix them. (6fd47db)
• [x86-64][s390x] mold now optimizes thread-local variable accesses in shared libraries if the library is linked with -z nodlopen. If your shared library is not intended to be used via dlopen(2) and your library frequently accesses thread-local variables, you might want to pass that option when linking your library. (25d02bb, f32ce33)
• [arm64] mold is now able to optimize GOT load by rewriting an ADDR+LDR instruction pair with an ADDR+ADD if the loaded GOT value is known at link-time. (f2311b1)

Bug fixes and compatibility improvements

• mold 1.9.0 was up to 10% slower than 1.8.0 on some multicore machines. We fixed the performance regression and made it even faster than 1.8.0. (7132822)
• Previously, mold failed to report an undefined symbol error if there's a weak undefined symbol of the same name. That bug resulted in producing a non-working executable instead of reporting a link failure. Now, mold correctly reports such link errors. (8936194)
• mold 1.9.0 might crash with SIGSEGV if --emit-relocs is used with object files containing debug info. That bug has been fixed. (e17d7da)

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• Uber Open Source
• Wei Wu
• G-Research-OSS
• 300baud
• Signal Slot Inc.
• Mercury
• Johan Andersson
• kyle-elliott
• Jinkyu Yi
• nasa
• Luke Kasz
• daquexian
• Brett Slatkin
• Dougall Johnson
• Santiago Pastorino
• JoJo
• Javier Blazquez
• Jonathan Goodman
• Markus Schirp

mold 1.9.0

New features

• mold gained support for the three new targets: 32-bit PowerPC, SH-4 and DEC Alpha. Each porting work didn't take more than a few days for us to complete, which demonstrate how portable the mold linker is. You can typically port mold to a new target just by writing a few hundreds lines of target-specific code. See arch-*.cc files in mold/elf/ directory to see how target-specific code actually looks like. (651adad, 3411e17, 6231510)

Bug fixes and compatibility improvements

• In a rare occasion, a statically-initialized function pointer might get a wrong address in a statically-linked executable. This bug has been fixed. (ccd47db)
• Fixed a -gdb-index option's crash bug on big-endian hosts. (3c96828)
• [RISC-V] mold rewrote machine instructions in a wrong way as a result of a wrong R_RISCV_HI20 relaxation if the output file was being linked against the high address. It's not a problem for user-land programs, but kernels linked with mold could crash due to this bug. This bug has been fixed. (3c96828)

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• Wei Wu
• G-Research-OSS
• Uber Open Source
• 300baud
• Signal Slot Inc.
• Johan Andersson
• Mercury
• kyle-elliott
• Jinkyu Yi
• Luke Kasz
• Santiago Pastorino
• nasa
• daquexian
• Markus Schirp
• Brett Slatkin
• Dougall Johnson
• JoJo
• Javier Blazquez
• Jonathan Goodman

mold 1.8.0

mold 1.8.0 is a new release of the high-speed linker.

New features

• The --relocatable (or -r) option has been reimplemented to improve its performance and compatibility with the GNU linkers. That option tells the linker to combine input object files into another object file instead of into an executable or a shared library file. mold has been supporting the feature since version 0.9, but until now the output file created with -r looked fairly different from what GNU linkers would produce. GHC (Glasgow Haskell Compiler) in particular uses re-linkable object files as dynamic libraries instead of real .so files, and it didn't work with mold. Now, mold can produce object files that GHC can load. Note that this work was funded by Mercury, so thanks to the company to help us improve the product. (Yes, you can ask us to prioritize your feature request by funding the project.) (c9a7ae7)
• --relocatable-merge-sections option has been added. By default, mold keeps original input section names for the --relocatable output and therefore does not merge input sections into a single output sections unless they are of the same name. If --relocatable-merge-sections is given, mold merges input by the usual default merging rule. For example, .text.foo and .text.bar are merged to .text if and only if --relocatable-merge-sections is given for the --relocatable output. (c2a0ae1)
• -z [no]dynamic-undefined-weak options have been added. This option controls whether an undefined weak symbol is promoted to a dynamic symbol or not. (ed235f3)
• --[no-]undefined-version options have been supported. Now, mold warns on a symbol name in a version script if it does not match with any defined symbol. This change was made so that it is easy to find a typo in a version script. (e2d7353)
• mold now warns on symbol type mismatch. If two object files have the same symbol with different symbol types, it usually means your program has a bug. Chances are, you are using the same identifier as a function name in one translation unit and as a global variable name in another. So it makes sense to warn on the mismatch. (b70211e)
• mold now merges .gnu.note.property sections for various x86 properties. (d30d743)

Removed features

• The experimental macOS/iOS support has been removed from mold. If you want to use it, please use our sold linker instead.

Bug fixes and compatibility improvements

• --wrap now works with LTO. (07d8911)
• A global variable initialized with an IFUNC function pointer is now initialized correctly with the function's address. Previously, it was mistakenly initialized to the function resolver's address. (b2858d2)
• The filename specified by --version-script or --dynamic-list is now searched from library search paths if it does not exist in the current working directory. This behavior is compatible with GNU linkers. (3c1a055, 8c87f16)
• mold now tries to avoid creating copy relocations as much as possible. This change fixed a compatibility issue with GHC. (5866c9e)
• Thread-local variables are now correctly aligned even if there's a TLV with a large alignment. (bd46edf)
• mold can now handle GCC LTO files created with -ffat-lto-objects. (804b843)
• mold now accepts -z nopack-relative-relocs as an alias for --pack-dyn-relocs=none for the sake of compatibility with GNU linkers. (b510588)
• mold now recognizes -z start-stop-visibility=hidden but ignores it because it's the default for mold. GNU linkers support this option to control the visibility of linker-synthesized __start_<sectname> and __stop_<sectname> symbols, with global as the default visibility. mold creates these symbols with the hidden visibility by default, which is desirable for almost all cases. (22c9ec8)
• [ARM32, i386] mold now emits REL-type relocations instead of RELA-type for the --relocatable output file. (8b373d3)

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• G-Research-OSS
• Uber Open Source
• Confiared S.R.L.
• Jim Huang
• 300baud
• Johan Andersson
• Signal Slot Inc.
• Wasmer
• Mercury
• kyle-elliott
• Johan Andersson
• Signal Slot Inc.
• Mercury
• kyle-elliott
• Johan Andersson
• windenntw
• disktnk
• Luke Kasz
• Dougall Johnson
• nasa
• H.T
• Brett Slatkin
• JoJo
• Javier Blazquez
• daquexian
• Jonathan Goodman
• Markus Schirp
• Luke Kasz
• Santiago Pastorino
• nasa
• Dougall Johnson
• Brett Slatkin
• JoJo
• Javier Blazquez
• daquexian
• Jonathan Goodman

mold 1.7.1

mold 1.7.1 is a bug-fix release containing only the following change:

Bug fix

• mold 1.7.0 may generate the same build-id for two different output files. We fixed the issue in 1.7.1 so that build-id is guaranteed to be unique for each different output file. (d8dd124)

Licensing

My comment in the last release notes about a possible license change caused an overwhelming response. Thank you guys for taking care of the software and its ecosystem. We will reconsider our plan based on the feedback. We may still want to change the license of mold/macOS, but we are not going to change the license of mold/Unix at least in a next few releases.

On this occasion, I want to say something. We are not a big evil corp who are trying to squeeze as much money as possible from users. This is mostly a one-person project, and what we are trying to do is to create better tools and make them publicly available to improve programmers' productivity worldwide. If you think of the number of developers who are using compiled languages and how many person-minutes we can save every year with better tools, the sum is a huge saving. We'd like to get a small chunk of it as a return. I believe we are doing good job at creating better tools but struggling to establish a way to get a return from it.

Please keep in mind that there are always people behind an open-source project. Some feedback to my comment were honestly too harsh and disrespectful. Open-source is as much as about people as it is about software. Please respect each other even if you have a different opinion.

By the way, for those who wish to obtain a copy of the mold linker in a different license than AGPL, we finally set up our company web site. You can purchase a license of the "sold" linker (which is a rebranded mold linker) using credit card. Unlike mold, sold is available under a usual per-user, per-month license. We believe this is a good option for some organizations, so please visit our website at bluewhale.systems to check it out.

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• Uber Open Source
• Confiared S.R.L.
• Jim Huang
• John Paul Adrian Glaubitz
• kyle-elliott
• Mercury
• Wasmer
• BuildBuddy
• windenntw
• haukevb
• Kurt Mahan
• Thomas Müller
• disktnk
• David Brownlee
• James Le Cuirot
• neoman/titan
• Carsten Strotmann
• Krzysztof Wolf
• Javier Blazquez
• JoJo
• Santiago Pastorino
• Dougall Johnson
• Brett Slatkin

mold 1.7.0

mold 1.7.0 is a new release of the high-speed linker.

Just like previous versions, you need to apply uxlfoundation/oneTBB@137c1a8 to libtbb if you do not use a bundled version of OneTBB library. OneTBB has merged this patch, but the most recent release of OneTBB hasn't picked it up yet.

Licensing

I'd like to inform users that I'm seriously considering changing the mold's license from AGPL to a source-available license unless I secure big funding. The new license would be something like individuals can use it for free but corporate users have to pay. mold started as my personal project, and I've been working on this full time for two years so far. I thought that I could earn a comfortable income if mold become popular, but unfortunately, I'm still losing my money. I think I need to take an action to make the project sustainable long term. For the details, please read my post.

New features

• [m68k] mold now supports the Motorola 68000 series microprocessors. Yes, it's the processor in the original Mac or Sun workstations in the 80s. This work is sponsored by m68k hobbyist communities.

Bug fixes and compatibility improvements

• We fixed a few issues for Facebook/Meta's BOLT optimizer (#789). Starting from the next LLVM release (we need llvm/llvm-project@20204db), BOLT should work on mold-generated executables out of the box.
• We fixed a long-standing symbol resolution issue involving GNU UNIQUE symbols which caused a link failure for a few programs. (730e970)
• Previously, if a version script contains a "C++" directive, and a symbol matches a non-C++ version pattern and a C++ version pattern, a wrong version could be assigned to the symbol. This has been fixed so that the mold's behavior matches with GNU ld. (9875150)

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• John Paul Adrian Glaubitz
• Mercury
• Wasmer
• BuildBuddy
• windenntw
• haukevb
• Kurt Mahan
• Thomas Müller
• disktnk
• David Brownlee
• James Le Cuirot
• neoman/titan
• Carsten Strotmann
• Krzysztof Wolf
• Santiago Pastorino
• Dougall Johnson
• Brett Slatkin
• Anonymous sponsor

mold 1.6.0

mold 1.6.0 is a new release of the high-speed linker. This release adds support for two IBM-based platforms, though we are not affiliated with IBM. We are happy to take donations and/or make support contracts. If you are interested in financially supporting the project, please visit our GitHub Sponsors page.

New features

• [ppc64] mold now supports the original 64-bit big-endian PowerPC ABI (which is also known as PPC64 ELFv1 or just ppc64), so that you can build applications for older PPC64 systems with mold. Note that this should not be confused with the modern PPC64 ELFv2 ABI (which is also known as ppc64le), which is already supported by mold.
• [s390x] Linux/s390x is now supported. Linux/s390x is the Linux environment on IBM z/Architecture mainframes. I've personally never seen a mainframe, but we wanted to support it because many Linux distros actively support that target, which in turn means there are many enterprise users who are using IBM mainframes. Speaking of the porting effort, we do not only port our linker to s390x but also found a couple of issues with the existing GCC toolchain for s390x. So, we are improving the whole IBM mainframe ecosystem!
• mold now creates smaller output files. It is most noticeable on targets with large page sizes such as PPC64 (on which the common page size is 64 KiB), but even on x86-64, it should save a few kilobytes per an output file.

Bug fixes and compatibility improvements

• [arm64] mold can now link executables with -static-pie. Previously, executables linked with that flag crashed immediately. (fc66759)

Acknowledgements

mold is an open-source project, and we accept donations via GitHub Sponsors and OpenCollective. We thank you to everybody who sponsors our project. In particular, we'd like to acknowledge the following organizations and people who have sponsored $32/mo or more during this release cycle:

• 300baud
• Mercury
• signal-slot
• Wei Wu
• BuildBuddy
• Luke Kasz
• Will Bond
• H.T
• .-A-.
• nasa
• haze
• daquexian
• Johan Andersson
• Markus Schirp
• Anonymous sponsor

mold 1.6.0-pre.1

This is a pre-release for those who want to test it before the official 1.6.0 release.

New features

• [ppc64] mold now supports 64-bit big-endian PowerPC ABI, which is also known as PPC64 ELFv1 or just ppc64. Older PowerPC systems using the processors in the big-endian mode are based on this ABI. Note that modern little-endian PowerPC systems are based on PPC64 ELFv2 ABI (which is also known as ppc64le), which is already supported by mold.
• [s390x] Linux/s390x is now supported. Linux/s390x is the Linux environment running on IBM z/Architecture mainframes. We do not only port our linker to s390x but also found a couple of issues in the existing GCC toolchain for s390x. So, we are improving the whole IBM mainframe ecosystem!
• mold now creates smaller output files. It is most noticeable on targets with large page sizes such as PPC64 (on which the common page size is 64 KiB), but even on x86-64, it should save a few kilobytes per an output file.

Bug fixes and compatibility improvements

• [arm64] mold can now link executables with -static-pie. Previously, executables linked with that flag crashed immediately.