mirror
/
rustc-dev-guide
mirror of https://github.com/rust-lang/rustc-dev-guide.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Address review comments

This commit is contained in:
Langston Barrett 2023-03-16 16:56:35 -04:00 committed by Michael Goulet
parent 34d85e1af8
commit b17ff13083
1 changed files with 9 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
src/fuzzing.md
View File

@ -26,9 +26,10 @@ project, please read this guide before reporting fuzzer-generated bugs!
*Please don't:* *Please don't:*
- Report lots of bugs that use internal features, including but not limited to - Don't report lots of bugs that use internal features, including but not
`custom_mir`, `lang_items`, `no_core`, and `rustc_attrs`. limited to `custom_mir`, `lang_items`, `no_core`, and `rustc_attrs`.
- Seed your fuzzer with inputs that are known to crash rustc (details below). - Don't seed your fuzzer with inputs that are known to crash rustc (details
below).
### Discussion ### Discussion
@ -107,16 +108,17 @@ these tools, post both the complete and minimized test cases. Generally,
## Effective fuzzing ## Effective fuzzing
When fuzzing rustc, you may want to avoid generating code, since this is mostly When fuzzing rustc, you may want to avoid generating machine code, since this
done by LLVM. Try `--emit=mir` instead. is mostly done by LLVM. Try `--emit=mir` instead.
A variety of compiler flags can uncover different issues. `-Zmir-opt-level=4` A variety of compiler flags can uncover different issues. `-Zmir-opt-level=4`
will turn on MIR optimization passes that are not run by default, potentially will turn on MIR optimization passes that are not run by default, potentially
uncovering interesting bugs. uncovering interesting bugs. `-Zvalidate-mir` can help uncover such bugs.
If you're fuzzing a compiler you built, you may want to build it with `-C If you're fuzzing a compiler you built, you may want to build it with `-C
target-cpu=native` or even PGO/BOLT to squeeze out a few more executions per target-cpu=native` or even PGO/BOLT to squeeze out a few more executions per
second. second. Of course, it's best to try multiple build configurations and see
what actually results in superior throughput.
## Existing projects ## Existing projects