ziglang/zig
260
6.0k
Fork
You've already forked zig
772

does all of compiler_rt conform to the zig-libc-test math tests? #30659

Open
opened 2026年01月02日 21:08:23 +01:00 by mercenary · 8 comments
Contributor
Copy link

Zig Version

0.16.0-dev.1871+eab93d357

Edited post (2026年05月04日)

As the following (specific) issue has been fixed, we should consider whether such an issue will occur again when removing more libc symbols. It may be smart to run zig-libc-test against the rest of compiler_rt.

Steps to Reproduce and Observed Behavior

First, compile a version of Zig which uses the compiler_rt implementation for some of the libc math functions. For example (edit: this branch has been fixed, so you should break it in another way):

git fetch -u https://codeberg.org/mercenary/zig 2879-rt:mercenary-2879-rt
git switch mercenary-2879-rt
stage3/bin/zig build -p stage4 -Dno-lib -Denable-llvm

Then, test stage4 against zig-libc-test:

zig="$(realpath stage4/bin/zig)"
cd /tmp/zig-libc-test
"$zig" build test -Dtarget=x86_64-linux-musl

The output is like this (edited for readability):

test
+- run exe log2 failure
error: ========= expected this stdout: =========
 ========= but found: ====================
 /cache/p/#/src/math/sanity/log2.h:1: bad fp exception: RN log2(-0x1.02239f3c6a8f1p+3)=nan, want INVALID got 0
 /cache/p/#/src/math/sanity/log2.h:3: bad fp exception: RN log2(-0x1.0c34b3e01e6e7p+3)=nan, want INVALID got 0
 /cache/p/#/src/math/sanity/log2.h:4: bad fp exception: RN log2(-0x1.a206f0a19dcc4p+2)=nan, want INVALID got 0
 /cache/p/#/src/math/sanity/log2.h:7: bad fp exception: RN log2(-0x1.a05cc754481d1p-2)=nan, want INVALID got 0
 /cache/p/#/src/math/sanity/log2.h:10: bad fp exception: RN log2(-0x1.5b86ea8118a0ep-1)=nan, want INVALID got 0
 /cache/p/#/src/math/special/log2.h:1: bad fp exception: RN log2(0x0p+0)=-inf, want DIVBYZERO got 0
 /cache/p/#/src/math/special/log2.h:2: bad fp exception: RN log2(-0x0p+0)=-inf, want DIVBYZERO got 0
 /cache/p/#/src/math/special/log2.h:3: bad fp exception: RN log2(-0x1p-100)=nan, want INVALID got 0
 /cache/p/#/src/math/special/log2.h:5: bad fp exception: RN log2(-0x1p+0)=nan, want INVALID got 0
 /cache/p/#/src/math/special/log2.h:7: bad fp exception: RN log2(-inf)=nan, want INVALID got 0
failed command: ./.zig-cache/o/1f8d705f5f07db37694239e72c4538c5/log2
test
+- run exe exp failure
error: ========= expected this stdout: =========
 ========= but found: ====================
 /cache/p/#/src/math/crlibm/exp.h:42: bad fp exception: RN exp(0x1.62e42fefa39fp+9)=inf, want INEXACT|OVERFLOW got 0
 /cache/p/#/src/math/crlibm/exp.h:44: bad fp exception: RU exp(0x1.62e42fefa39fp+9)=inf, want INEXACT|OVERFLOW got 0
 /cache/p/#/src/math/crlibm/exp.h:46: bad fp exception: RD exp(0x1.62e42fefa39fp+9)=0x1.fffffffffffffp+1023, want INEXACT|OVERFLOW got 0
 /cache/p/#/src/math/crlibm/exp.h:48: bad fp exception: RZ exp(0x1.62e42fefa39fp+9)=0x1.fffffffffffffp+1023, want INEXACT|OVERFLOW got 0
 /cache/p/#/src/math/crlibm/exp.h:49: bad fp exception: RN exp(-0x1.74910d52d3051p+9)=0x1p-1074, want INEXACT|UNDERFLOW got INEXACT
[SNIP]
 /cache/p/#/src/math/sanity/log10.h:7: bad fp exception: RN log10(-0x1.a05cc754481d1p-2)=nan, want INVALID got 0
 /cache/p/#/src/math/sanity/log10.h:10: bad fp exception: RN log10(-0x1.5b86ea8118a0ep-1)=nan, want INVALID got 0
 /cache/p/#/src/math/special/log10.h:1: bad fp exception: RN log10(0x0p+0)=-inf, want DIVBYZERO got 0
 /cache/p/#/src/math/special/log10.h:2: bad fp exception: RN log10(-0x0p+0)=-inf, want DIVBYZERO got 0
 /cache/p/#/src/math/special/log10.h:3: bad fp exception: RN log10(-0x1p-100)=nan, want INVALID got 0
 /cache/p/#/src/math/special/log10.h:5: bad fp exception: RN log10(-0x1p+0)=nan, want INVALID got 0
 /cache/p/#/src/math/special/log10.h:7: bad fp exception: RN log10(-inf)=nan, want INVALID got 0
failed command: ./.zig-cache/o/1926366952ad1c89b1a1f593c3d14688/log10
Build Summary: 501/511 steps succeeded (9 failed)
test transitive failure
+- run exe acoshf failure
+- run exe exp failure
+- run exe lgamma_r failure
+- run exe log10 failure
+- run exe log10f failure
+- run exe log2 failure
+- run exe log2f failure
+- run exe log failure
+- run exe logf failure

This issue blocks ziglang/zig!30648, which tries to remove libc functions in favor of the existing compiler_rt implementations, contributing to #2879. See also this comment in the PR.

Expected Behavior

I expect the compiler_rt implementations not to fail any additional tests, as compared to the musl implementations.

### Zig Version 0.16.0-dev.1871+eab93d357 ### Edited post (2026年05月04日) As the following (specific) issue has been fixed, we should consider whether such an issue will occur again when removing more _libc_ symbols. It [may be smart](https://codeberg.org/ziglang/zig/issues/30659#issuecomment-9541586) to run zig-libc-test against the rest of _compiler_rt_. ### Steps to Reproduce and Observed Behavior First, compile a version of Zig which uses the _compiler_rt_ implementation for some of the _libc_ math functions. For example (**edit**: this branch has been fixed, so you should break it in another way): ```sh git fetch -u https://codeberg.org/mercenary/zig 2879-rt:mercenary-2879-rt git switch mercenary-2879-rt stage3/bin/zig build -p stage4 -Dno-lib -Denable-llvm ``` Then, test stage4 against [zig-libc-test](https://github.com/rpkak/zig-libc-test): ```sh zig="$(realpath stage4/bin/zig)" cd /tmp/zig-libc-test "$zig" build test -Dtarget=x86_64-linux-musl ``` The output is like this (edited for readability): ``` test +- run exe log2 failure error: ========= expected this stdout: ========= ========= but found: ==================== /cache/p/#/src/math/sanity/log2.h:1: bad fp exception: RN log2(-0x1.02239f3c6a8f1p+3)=nan, want INVALID got 0 /cache/p/#/src/math/sanity/log2.h:3: bad fp exception: RN log2(-0x1.0c34b3e01e6e7p+3)=nan, want INVALID got 0 /cache/p/#/src/math/sanity/log2.h:4: bad fp exception: RN log2(-0x1.a206f0a19dcc4p+2)=nan, want INVALID got 0 /cache/p/#/src/math/sanity/log2.h:7: bad fp exception: RN log2(-0x1.a05cc754481d1p-2)=nan, want INVALID got 0 /cache/p/#/src/math/sanity/log2.h:10: bad fp exception: RN log2(-0x1.5b86ea8118a0ep-1)=nan, want INVALID got 0 /cache/p/#/src/math/special/log2.h:1: bad fp exception: RN log2(0x0p+0)=-inf, want DIVBYZERO got 0 /cache/p/#/src/math/special/log2.h:2: bad fp exception: RN log2(-0x0p+0)=-inf, want DIVBYZERO got 0 /cache/p/#/src/math/special/log2.h:3: bad fp exception: RN log2(-0x1p-100)=nan, want INVALID got 0 /cache/p/#/src/math/special/log2.h:5: bad fp exception: RN log2(-0x1p+0)=nan, want INVALID got 0 /cache/p/#/src/math/special/log2.h:7: bad fp exception: RN log2(-inf)=nan, want INVALID got 0 failed command: ./.zig-cache/o/1f8d705f5f07db37694239e72c4538c5/log2 test +- run exe exp failure error: ========= expected this stdout: ========= ========= but found: ==================== /cache/p/#/src/math/crlibm/exp.h:42: bad fp exception: RN exp(0x1.62e42fefa39fp+9)=inf, want INEXACT|OVERFLOW got 0 /cache/p/#/src/math/crlibm/exp.h:44: bad fp exception: RU exp(0x1.62e42fefa39fp+9)=inf, want INEXACT|OVERFLOW got 0 /cache/p/#/src/math/crlibm/exp.h:46: bad fp exception: RD exp(0x1.62e42fefa39fp+9)=0x1.fffffffffffffp+1023, want INEXACT|OVERFLOW got 0 /cache/p/#/src/math/crlibm/exp.h:48: bad fp exception: RZ exp(0x1.62e42fefa39fp+9)=0x1.fffffffffffffp+1023, want INEXACT|OVERFLOW got 0 /cache/p/#/src/math/crlibm/exp.h:49: bad fp exception: RN exp(-0x1.74910d52d3051p+9)=0x1p-1074, want INEXACT|UNDERFLOW got INEXACT [SNIP] /cache/p/#/src/math/sanity/log10.h:7: bad fp exception: RN log10(-0x1.a05cc754481d1p-2)=nan, want INVALID got 0 /cache/p/#/src/math/sanity/log10.h:10: bad fp exception: RN log10(-0x1.5b86ea8118a0ep-1)=nan, want INVALID got 0 /cache/p/#/src/math/special/log10.h:1: bad fp exception: RN log10(0x0p+0)=-inf, want DIVBYZERO got 0 /cache/p/#/src/math/special/log10.h:2: bad fp exception: RN log10(-0x0p+0)=-inf, want DIVBYZERO got 0 /cache/p/#/src/math/special/log10.h:3: bad fp exception: RN log10(-0x1p-100)=nan, want INVALID got 0 /cache/p/#/src/math/special/log10.h:5: bad fp exception: RN log10(-0x1p+0)=nan, want INVALID got 0 /cache/p/#/src/math/special/log10.h:7: bad fp exception: RN log10(-inf)=nan, want INVALID got 0 failed command: ./.zig-cache/o/1926366952ad1c89b1a1f593c3d14688/log10 Build Summary: 501/511 steps succeeded (9 failed) test transitive failure +- run exe acoshf failure +- run exe exp failure +- run exe lgamma_r failure +- run exe log10 failure +- run exe log10f failure +- run exe log2 failure +- run exe log2f failure +- run exe log failure +- run exe logf failure ``` This issue blocks ziglang/zig!30648, which tries to remove _libc_ functions in favor of the existing _compiler_rt_ implementations, contributing to [#2879](https://github.com/ziglang/zig/issues/2879). See also [this comment in the PR](https://codeberg.org/ziglang/zig/pulls/30648#issuecomment-9455507). ### Expected Behavior I expect the _compiler_rt_ implementations not to fail any additional tests, as compared to the _musl_ implementations.
Contributor
Copy link

I created a branch, which contains some changes related to this.

I created a [branch](https://codeberg.org/rpkak/zig/src/branch/libc-compiler_rt-math), which contains some changes related to this.
andrewrk added this to the Urgent milestone 2026年01月02日 22:05:02 +01:00
Author
Contributor
Copy link

Is there an easy way to prefer compiler_rt symbols over libc symbols? That would make it possible to directly test compiler_rt without ripping out libc files first, to easily see whether the existing compiler_rt symbols comply with the tests.

Is there an easy way to prefer _compiler_rt_ symbols over _libc_ symbols? That would make it possible to directly test _compiler_rt_ without ripping out _libc_ files first, to easily see whether the existing _compiler_rt_ symbols comply with the tests.

The specific test failures referenced in this issue were addressed in the pull request it was raised before the PR was merged.

For the sake of clarity, does this issue correspond to the small set of math tests that were already fixed, or does it correspond to investigating and fixing all the math tests that are still considered unstable and commented out in the libc test suite?

The specific test failures referenced in this issue were [addressed](https://codeberg.org/ziglang/zig/pulls/30648/commits/81097669270c9f525eb0556915bdbcc3492fcec3) in the [pull request](https://codeberg.org/ziglang/zig/pulls/30648) it was raised before the PR was merged. For the sake of clarity, does this issue correspond to the small set of math tests that were already fixed, or does it correspond to investigating and fixing all the math tests that are still [considered unstable](https://github.com/rpkak/zig-libc-test/blob/main/build.zig) and [commented out](https://codeberg.org/ziglang/zig/src/commit/b94a65ed403f440a53ae29f9f3cc9643fd2076d9/test/libc.zig) in the libc test suite?
Author
Contributor
Copy link

I'd say the PR only fixed the part of compiler_rt that was newly used by the patch. The rest of compiler_rt may still be non-compliant. See my previous comment on what comes next.

I'd say the PR only fixed the part of compiler_rt that was newly used by the patch. The rest of compiler_rt may still be non-compliant. See my previous comment on what comes next.
mercenary changed title from (削除) compiler_rt does not conform to the zig-libc-test math tests (削除ここまで) to does all of compiler_rt conform to the zig-libc-test math tests? 2026年05月04日 08:49:21 +02:00

Looking through the commented-out math tests in test/libc.zig, it looks to me as though only fma has an implementation in compiler_rt. The remaining commented-out tests are due to the existing libc implementations failing their respective libc tests.

Should this issue could be interpreted in two ways: it corresponds to some compiler_rt functions whose implementations failed some libc-test tests, but snuck through before test-libc-nsz was added, or it's a corollary issue to issue 30978, and should remain open as long as there are math functions to port to compiler_rt. Which is the correct way to interpret this issue?

The rest of compiler_rt may still be non-compliant

I'm leaning towards the former interpretation because of this statement - in which case, I believe that only the fma functions are non-compliant. Although that implementation was added after this issue was created.

Looking through the [commented-out math tests in test/libc.zig](https://codeberg.org/ziglang/zig/src/branch/master/test/libc.zig), it looks to me as though only fma has [an implementation in compiler_rt](https://codeberg.org/ziglang/zig/src/branch/master/lib/compiler_rt/fma.zig). The remaining commented-out tests are due to the existing libc implementations failing their respective libc tests. Should this issue could be interpreted in two ways: it corresponds to some compiler_rt functions whose implementations failed some libc-test tests, but snuck through before `test-libc-nsz` was added, or it's a corollary issue to [issue 30978](https://codeberg.org/ziglang/zig/issues/30978), and should remain open as long as there are math functions to port to compiler_rt. Which is the correct way to interpret this issue? > The rest of compiler_rt may still be non-compliant I'm leaning towards the former interpretation because of this statement - in which case, I believe that only the `fma` functions are non-compliant. Although that implementation was added after this issue was created.
Contributor
Copy link

Also the compiler_rt implementations of expl, exp2l, logl, log2l and log10l, fail the test-libc-nsz. The reason, why these tests are not commented out, is that for these functions, the musl implementation is still used.

Also the compiler_rt implementations of `expl`, `exp2l`, `logl`, `log2l` and `log10l`, fail the `test-libc-nsz`. The reason, why these tests are not commented out, is that for these functions, the musl implementation is still used.

I've created a branch to test the failed tests you've described. I'm reproducing the instructions from the original post using my own branch and zig build test-libc-nsz, but I don't seem to be able to get the logl family of tests to fail. I'm able to reproduce the test failures of the expl tests, though. Is my approach to reproducing these test failures correct?

I've [created a branch](https://codeberg.org/squingo44/zig/commit/2f088a3cbf398ee5b9a867dbaa82ede2a7c7820c#diff-e78cd25dae9d3d3ffc7a360e89a5a8b129351fbe) to test the failed tests you've described. I'm reproducing the instructions from the original post using my own branch and `zig build test-libc-nsz`, but I don't seem to be able to get the `logl` family of tests to fail. I'm able to [reproduce the test failures](https://zigbin.io/5b5677) of the `expl` tests, though. Is my approach to reproducing these test failures correct?
Contributor
Copy link

You need to also comment out files like musl/src/math/x86_64/logl.s.

You need to also comment out files like `musl/src/math/x86_64/logl.s`.
Sign in to join this conversation.
No Branch/Tag specified
master
elfv2
spork8
restricted
0.16.x
panic-rewrite
parking-futex-lockfree
windows-Io-cleanup
Io-watch
ParseCommandLineOptions
windows-async-files
poll-ring
debug-file-leaks-differently
debug-file-leaks
ProcessPrng
elfv2-dyn
jobserver
threadtheft
io-threaded-no-queue
0.15.x
Io.net
comptime-allocator
restricted-function-pointers
cli
wasm-linker-writer
wrangle-writer-buffering
sha1-stream
async-await-demo
fixes
0.14.x
ast-node-methods
macos-debug-info
make-vs-configure
fuzz-macos
sans-aro
ArrayList-reserve
incr-bug
llvm-ir-nosanitize-metadata
ci-tarballs
ci-scripts
threadpool
0.12.x
new-pkg-hash
json-diagnostics
more-doctests
rework-comptime-mutation
0.11.x
ci-perf-comment
stage2-async
0.10.x
autofix
0.9.x
aro
hcs
0.8.x
0.7.x
0.16.0
0.15.2
0.15.1
0.15.0
0.14.1
0.14.0
0.12.1
0.13.0
0.12.0
0.11.0
0.10.1
0.10.0
0.9.1
0.9.0
0.8.1
0.8.0
0.7.1
0.7.0
0.6.0
0.5.0
0.4.0
0.3.0
0.2.0
0.1.1
0.1.0
Labels
Clear labels
abi/f32
abi/ilp32
abi/sf
accepted
This proposal is planned.
arch/21k
arch/6502
arch/aarch64
arch/alpha
arch/amdgcn
arch/arc
arch/arc32
arch/arc64
arch/arm
arch/avr
arch/bfin
arch/bpf
arch/colossus
arch/cris
arch/csky
arch/dlx
arch/epiphany
arch/fr30
arch/frv
arch/hexagon
arch/hppa
arch/hppa64
arch/ia64
arch/kalimba
arch/kvx
arch/lanai
arch/lm32
arch/loongarch32
arch/loongarch64
arch/m32r
arch/m68k
arch/m88k
arch/mcore
arch/microblaze
arch/mips
arch/mips64
arch/mmix
arch/moxie
arch/mrisc32
arch/msp430
arch/nds32
arch/ns32k
arch/nvptx
arch/or1k
arch/powerpc
arch/powerpc64
arch/propeller
arch/riscv32
arch/riscv64
arch/rl78
arch/rx
arch/s390x
arch/sh
arch/sparc
arch/sparc64
arch/spirv
arch/spu
arch/tricore
arch/v850
arch/vax
arch/vc4
arch/ve
arch/wasm
arch/x86
arch/x86_64
arch/xcore
arch/xtensa
autodoc
The web application for interactive documentation and generation of its assets.
backend/c
The C backend outputs C source code.
backend/llvm
The LLVM backend outputs an LLVM bitcode module.
backend/self-hosted
The self-hosted backends produce machine code directly.
binutils
Zig's included binary utilities: zig ar, zig dlltool, zig lib, zig ranlib, zig objcopy, and zig rc.
breaking
Implementing this issue could cause existing code to no longer compile or have different behavior.
build system
The Zig build system - zig build, std.Build, the build runner, and package management.
debug info
An issue related to debug information (e.g. DWARF) produced by the Zig compiler.
docs
An issue with documentation, e.g. the language reference or standard library doc comments.
error message
This issue points out an error message that is unhelpful and should be improved.
frontend
Tokenization, parsing, AstGen, ZonGen, Sema, Legalize, and Liveness.
fuzzing
An issue related to Zig's integrated fuzz testing.
incremental
Reuse of internal compiler state for faster compilation.
lib/c
This issue relates to Zig's libc implementation and/or vendored libcs.
lib/compiler-rt
This issue relates to Zig's compiler-rt library.
lib/cxx
This issue relates to Zig's vendored libc++ and/or libc++abi.
lib/std
This issue relates to Zig's standard library.
lib/tsan
This issue relates to Zig's vendored libtsan.
lib/ubsan-rt
This issue relates to Zig's ubsan-rt library.
lib/unwind
This issue relates to Zig's vendored libunwind.
linking
Zig's integrated object file and incremental linker.
miscompilation
The compiler reports success but produces semantically incorrect code.
os/android
os/contiki
os/dragonfly
os/driverkit
os/emscripten
os/freebsd
os/fuchsia
os/haiku
os/hermit
os/hurd
os/illumos
os/ios
os/linux
os/maccatalyst
os/macos
os/managarm
os/netbsd
os/ohos
os/openbsd
os/plan9
os/redox
os/rtems
os/serenity
os/tvos
os/uefi
os/visionos
os/wasi
os/watchos
os/windows
proposal
This issue suggests language modifications. If it also has the "accepted" label then it is planned.
release notes
This issue or pull request should be mentioned in the release notes.
testing
This issue is related to testing the compiler, standard library, or other parts of Zig.
zig cc
Zig as a drop-in C-family compiler.
zig fmt
The Zig source code formatter.
zig reduce
The Zig source code reduction tool.
bounty
https://ziglang.org/news/announcing-donor-bounties
bug
Observed behavior contradicts documented or intended behavior.
contributor-friendly
This issue is limited in scope and/or knowledge of project internals.
downstream
An issue with a third-party project that uses this project.
enhancement
Solving this issue will likely involve adding new logic or components to the codebase.
infra
An issue related to project infrastructure, e.g. continuous integration.
optimization
A task to improve performance and/or resource usage.
question
No questions on the issue tracker; use a community space instead.
regression
Something that used to work in a previous version stopped working
upstream
An issue with a third-party project that this project uses.
use case
Describes a real use case that is difficult or impossible, but does not propose a solution.
No labels
abi/f32
abi/ilp32
abi/sf
accepted
arch/21k
arch/6502
arch/aarch64
arch/alpha
arch/amdgcn
arch/arc
arch/arc32
arch/arc64
arch/arm
arch/avr
arch/bfin
arch/bpf
arch/colossus
arch/cris
arch/csky
arch/dlx
arch/epiphany
arch/fr30
arch/frv
arch/hexagon
arch/hppa
arch/hppa64
arch/ia64
arch/kalimba
arch/kvx
arch/lanai
arch/lm32
arch/loongarch32
arch/loongarch64
arch/m32r
arch/m68k
arch/m88k
arch/mcore
arch/microblaze
arch/mips
arch/mips64
arch/mmix
arch/moxie
arch/mrisc32
arch/msp430
arch/nds32
arch/ns32k
arch/nvptx
arch/or1k
arch/powerpc
arch/powerpc64
arch/propeller
arch/riscv32
arch/riscv64
arch/rl78
arch/rx
arch/s390x
arch/sh
arch/sparc
arch/sparc64
arch/spirv
arch/spu
arch/tricore
arch/v850
arch/vax
arch/vc4
arch/ve
arch/wasm
arch/x86
arch/x86_64
arch/xcore
arch/xtensa
autodoc
backend/c
backend/llvm
backend/self-hosted
binutils
breaking
build system
debug info
docs
error message
frontend
fuzzing
incremental
lib/c
lib/compiler-rt
lib/cxx
lib/std
lib/tsan
lib/ubsan-rt
lib/unwind
linking
miscompilation
os/android
os/contiki
os/dragonfly
os/driverkit
os/emscripten
os/freebsd
os/fuchsia
os/haiku
os/hermit
os/hurd
os/illumos
os/ios
os/linux
os/maccatalyst
os/macos
os/managarm
os/netbsd
os/ohos
os/openbsd
os/plan9
os/redox
os/rtems
os/serenity
os/tvos
os/uefi
os/visionos
os/wasi
os/watchos
os/windows
proposal
release notes
testing
zig cc
zig fmt
zig reduce
bounty
bug
contributor-friendly
downstream
enhancement
infra
optimization
question
regression
upstream
use case
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
3 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
ziglang/zig#30659
Reference in a new issue
ziglang/zig
No description provided.
Delete branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?