Stability improve #773

tokatoka · 2022-09-10T11:22:21Z

Before, state.stability() is about the stability of the last corpus entry that goes throught the calibration stage.
Now it keeps track of which unstable index and then send monitors the number of unstable indexes from all corpus divided by map size.

codecov-commenter · 2022-09-10T11:58:12Z

Codecov Report

Merging #773 (66aaf5f) into main (0fe8192) will increase coverage by 0.00%.
The diff coverage is 3.70%.

@@           Coverage Diff           @@
##             main     #773   +/-   ##
=======================================
  Coverage   14.63%   14.63%           
=======================================
  Files         150      150           
  Lines       17822    17828    +6     
=======================================
+ Hits         2609     2610    +1     
- Misses      15213    15218    +5

Impacted Files	Coverage Δ
libafl/src/events/mod.rs	`14.70% <0.00%> (-0.22%)`	⬇️
libafl/src/feedbacks/differential.rs	`59.32% <ø> (+1.94%)`	⬆️
libafl/src/fuzzer/mod.rs	`19.51% <ø> (ø)`
libafl/src/stages/mod.rs	`6.49% <ø> (ø)`
libafl/src/stages/push/mod.rs	`0.00% <ø> (ø)`
libafl/src/stages/push/mutational.rs	`0.00% <ø> (ø)`
libafl/src/state/mod.rs	`15.17% <ø> (+0.79%)`	⬆️
libafl_targets/src/cmplog.rs	`0.00% <ø> (ø)`
libafl/src/stages/calibrate.rs	`0.90% <4.54%> (+0.90%)`	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

tokatoka · 2022-09-10T12:00:05Z

@WorksButNotTested
Now it shows unstable entries gathered from every corpus entries..

libafl/src/stages/calibrate.rs

tsheinen · 2022-09-10T14:41:20Z

libafl/src/stages/calibrate.rs

+            } else {
+                state.add_metadata::<UnstableEntriesMetadata>(UnstableEntriesMetadata::new(
+                    HashSet::from_iter(unstable_entries),
+                    map_len,


is map_len what we want here? IMO (unstable_edges / discovered_edges) would be better than (unstable_edges / all_edges). As it is now undiscovered edges are assumed to be stable which can inflate the stability score, especially when the map is larger than the expected number of discovered edges.

thanks for the suggestion
we already update the broker with (discovered_edges / all_edges), I feel it is fine to show stability as (unstable_edges / all_edges) right next to it. like this

[Stats #7] (GLOBAL) run time: 0h-0m-31s, clients: 9, corpus: 8401, objectives: 0, executions: 2505102, exec/sec: 81319 (CLIENT) corpus: 1089, objectives: 0, executions: 307364, exec/sec: 10062, edges: 2205/2219 (99%), stability: 434/2179 (19%) [Stats #8] (GLOBAL) run time: 0h-0m-31s, clients: 9, corpus: 8401, objectives: 0, executions: 2505102, exec/sec: 81319 (CLIENT) corpus: 1098, objectives: 0, executions: 292780, exec/sec: 9618, edges: 2204/2219 (99%), stability: 417/2201 (18%)

Because these two numbers are calculated at different location, I have to either

In calibration stage, run through all the maps to count the non-0 element (but this will repeat the same computation done in MaxMapfeedback::is_interesting, which is wastful of cpu poweres)

In is_interesting, update write the number of discovered_edges number into the metadata in state, but this function is a super hot path, personally I would not like to add additional overhead here..

Yeah that makes sense. It isn't really the job of the calibration stage to compute that and the overhead doesn't seem worth it. It would probably be best to do it in the monitor if it's done at all? Because the monitor has all that data already. but that's kinda special-case-y and might be best to just have people modify a monitor if they want it to display like that. I do think that (unstable_edges / discovered_edges) is a more useful metric but that's admittedly rather subjective lol.

With the dynamically adjustable map size, the difference is likely to be minimal anyway. But since it isn’t really a super precise metric anyway, and is only intended to give a rough idea, it feels like it’s probably fine as is. In any case the user can calculate the original style score with some simple maths from the output. It would be helpful to document what the metric is showing exactly and perhaps how it differs to AFL++ as I imagine many users will be transitioning from it to LibAFL and may expect it to be exactly the same, so may want to account for any differences to build confidence in making the switch.

sure, I'll add docs tomorrow

Awesome. Thanks.

domenukk · 2022-09-11T09:44:57Z

libafl/src/events/mod.rs

@@ -377,13 +378,15 @@ where
                },
            )?;

-            if let Some(x) = state.stability() {
-                let stability = f64::from(*x);
+            if state.has_metadata::<UnstableEntriesMetadata>() {


This is an additional (quite costly) HashMap lookup and shouldn't be used if you will immediately get the value anyway

domenukk · 2022-09-11T10:02:41Z

libafl/src/stages/calibrate.rs

            {
                if *first != *cur && *history != O::Entry::max_value() {
                    *history = O::Entry::max_value();
-                    unstable_entries += 1;
+                    unstable_entries.insert(idx);


You could use a bool array of instead of a hashset for this step, should be a lot more cache friendly (hashmaps are not actually fast)

(An array won't always be faster but for a large amount of elements it might be?)
Alternatively, just use a Vec here, it's getting inserted in a set later, anyway.
Other idea, we could benchmark BTreeSet vs HashSet

Fwiw we can add micro benchmarks if we're interested, like so:
https://github.com/AFLplusplus/LibAFL/blob/2504b6dae3584f12b7ed1309919d6a5611278545/utils/libafl_benches/benches/rand_speeds.rs

If we are worried about performance, then another option might be to only update the stats every N seconds, or N iterations. It is unlikely someone will need instantaneous updates?

Calibration is done for every new testcase right now, there are other things associated with it

I meant maybe just the bit for stability, rather than the whole thing. But I don’t know the architecture of LibAFL well enough to say whether that’s practical.

ok i'll use vec

* initial * add * fmt & fix * dbg remove * clp * clp * more * clippy * del * fix * remove unused * fix * doc

* Mac OS Autotokens (#723) * mac_tokens * more * win fix * fmt * fmt c * Use nightly fmt (#728) * Fix compilation for aarch64 qemu (#731) Typo lead to fail to compile for arm64 * Simd Fix (#729) * simd fix * fmt * Fixing readme & docs (#730) * fix * fix * add * add * fmt * 0.8.1 (#732) * New Pass Manager Arguments (#724) * new pm arguments * enable abgeana's code * Fix tui with 1 client (#734) * unbreak tui with 1 client * clippy * Add core affinity support for FreeBSD (#736) * NYX Executor (GSoC '22) (#693) * Add ccache * Update codecov.yml * Add libnyx * Fix * Add nyx build script * Fix build.sh && init executor.rs * Fix commit * Fix code * initialize `exector.rs` * refine API in `nyx_bridge.rs` * initialze `run_target` * add `test_nyxhelper` * initize `test_executor` * remove `nyx_beidge.rs` * make `test_executor` compile * Improve test * refine code * update version * fix docker * fix docker * Fix clippy * Fix build * fix build && add `set_timeout` * Fix and refine CI * fix CI * Fix CI * Add platform restrict * cargo fmt * add parallel mode * add example `nyx_libxml2_parallel` * fix fuzzer example * fix CI * add README * fix CI * fix CI * fix CI * remove unwrap and NyxResult * code format fix * add libnyx's rev * fix format * change Duration format && Fix CI * caego fmt * fix CI * fix CI * Add doc * test CI * Update test_all_fuzzers.sh * Update test_all_fuzzers.sh * Update test_all_fuzzers.sh * add cache for apt and cargo-install * Update build_and_test.yml * Update build_and_test.yml * tmp test CI * fix CI * remove debug cmd * remove test * code refine * code refine * code refine * code refine * add Makefile * fix example doc for nyx * add `NyxHelper::new_with_initial_timeout` * fix `NyxHelper::new` * fix curl parameter * code refine * add check for setup script * use afl-clang-fast in nyx * fix logic * fix makefile * fix CI * Update build_and_test.yml * Update build_and_test.yml * remove debug cmd Co-authored-by: syheliel <syheliel@gmail.com> Co-authored-by: Dominik Maier <dmnk@google.com> * Fix spelling error (#745) * OSX force_load option (#743) * Update clang.rs * fmt * Add continous JSON Logging monitor (#738) * Add simple JSON Monitor * Add documentation * Log global state * Fix formatting * Save state depending on closure outcome, have file opened all the time * Make OnDiskJSONMonitor cloneable * Switch to FnMut to allow stateful closures * Use &mut M: Monitor for the closure * Fix documentation of Rand::below (#747) * Netopenbsd build fix (#746) * core affinity netbsd implementation. * openbsd build fix * Fix autotokens doc (#751) * fix * remove wrong doc * Simplification for netbsd-specific code (#750) the cpuset api is already present in libc... * Add test case minimising stage (tmin) (#735) * add test case minimising stage * general purpose minimiser impl, with fuzzer example * reorganise, document, and other cleanup * correct python API return value * correct some docs * nit: versioning in fuzzers * ise -> ize * Implement a corpus minimiser (cmin) (#739) * initial try * correct case where cull attempts to fetch non-existent corpus entries * various on_remove, on_replace implementations * ise -> ize (consistency), use TestcaseScore instead of rolling our own * oops, feature gate * documentation! * link c++ * doc-nit: correction in opt explanation don't write documentation at 0300 * better linking * Skippable stage, generator wrapper for Grimoire (#748) * Skippable stage, generator wrapper for Grimoire * more fancy wrapper * MapFeedback: Adding support for with_name() (#752) * Adding support for with_name() * Adding with_name() function description * dragonflybsd build fix for core affinity. (#753) supporting most of linux sched api here. * CI for FreeBSD (#754) * CI for FreeBSD * rustup -y? * fixed path, switched to clippy * bsd don't source * added llvm * clippy * more yml * ? * testing ci * llvm? * llvm?? * more llvm, more tests * fixed testcase' * mem limits * more sudo * reenable all the CI * Fixes for new Clippy (#755) * New Clippy fixes for QEMU (#757) * Core affinity for FreeBSD pinning task to the wanted cpu (#756) * Do not zero-init struct in QEMU (#758) * New Clippy fixes for QEMU * no need to 0-initialize mem * clippy * Add doc for libafl_nyx (#759) Co-authored-by: syheliel <syheliel@gmail.com> * Adjust NyxExecutor trait bound to HasTargetBytes from HasBytesVec (#760) * adjust NyxExecutor trait bound to HasTargetBytes from HasBytesVec * oops actually use HasTargetBytes instead * libafl_frida: ASan hook adding Apple's memset_pattern* api. (#761) * Fix cargo doc on windows (#762) * add doc cfg * fix nostd docs * ignore CommandConfigurator doc test execution on non-unix platform * add cargo doc step pipeline on windows platform * Enable memset_patter ASan hooks for Apple on libafl_frida (#763) * Fix forkserver options (#771) * Stability improve (#773) * initial * add * fmt & fix * dbg remove * clp * clp * more * clippy * del * fix * remove unused * fix * doc * Fix doc (#780) * Add track_stability option to CalibrationStage (#781) * add * Update gramatron.rs * Update emu.rs * try * clp * Dump registers on freebsd x86_64 (#779) * Illumos support (#775) implementing core affinity too. * Reduce clang warnings for version output in libafl_cc. (#778) * Extend gramatron recursive mutator (#783) * Dump registers on NetBSD amd64 (#786) * Add support for ARMBE8 (#768) * Changes to build QEMU out-of-tree so that we don't need to clone the repo for each feature combination we build * Add be support to libafl_qemu * More config tweaks Co-authored-by: Your Name <you@example.com> * [AFLplusplus/LibAFL] dump registers on OpenBSD amd64 (PR #787) * dump registers on openbsd * write_crash implementations * Windows gdiplus (#789) * Initial steps * Harness code cleanup * don't panic on linux in order not to break the CI * formatting once again * restored cfg unix to unbreak linux build * Remove clang download from windows CI (#791) * Attempt to remove clang 12 setup * frida_gdiplus added to CI * Gdiplus comments (#792) * Attempt to remove clang 12 setup * frida_gdiplus added to CI * Redundancy note * formatting again :\ * mistake of directory name * Fix len miscalculation in grimoire string replace (#794) * Fix len miscalculation in grimoire string replace * ok Rust i was writing JS these days Co-authored-by: Andrea Fioraldi <andrea.fioraldi@trellix.com> * Fix doc typos (#796) * Fix CI (#798) * bump (#799) * Support for write_crash on netbsd (#788) * Support for bolts::cpu::read_time_counter on arm64 (#790) * Add ability to use virtual dispatch to StagesTuple (#801) * Add ability to use virtual dispatch to stagesTuple * Fix lint * Adding CPSR register for arm qemu (#800) * trying to add in observer * writing test * got up to running with instrumentation but i still need to get the map * fixing fuzzer code * adding tinyinst fuzzer * adding ffi to store all the map data into vec. * adding some new things * adding somewhat state of how i would like it should work * fixing some things * alot of false positives. * fixing before adding args * updated to use FileInput! * adding build script to pull tinyinst * fixing git issue * writing instruction to run how to run tinyinst fuzzer Co-authored-by: Dongjia Zhang <tokazerkje@outlook.com> Co-authored-by: Dominik Maier <dmnk@google.com> Co-authored-by: Phan Thanh Duy <phanthanhduypr@gmail.com> Co-authored-by: Nicholas Lang <97475577+nicklangsysdig@users.noreply.github.com> Co-authored-by: David CARLIER <devnexen@gmail.com> Co-authored-by: syheliel <45957390+syheliel@users.noreply.github.com> Co-authored-by: syheliel <syheliel@gmail.com> Co-authored-by: Aiden Hall <AidenRHall@users.noreply.github.com> Co-authored-by: Sönke <eknoes@users.noreply.github.com> Co-authored-by: Sirui Mu <msrlancern@gmail.com> Co-authored-by: Addison Crump <me@addisoncrump.info> Co-authored-by: Patrick Gersch <gersch.patrick@gmail.com> Co-authored-by: Teddy Heinen <teddy@heinen.dev> Co-authored-by: Vincent <space_white@yahoo.com> Co-authored-by: Andrea Fioraldi <andreafioraldi@gmail.com> Co-authored-by: WorksButNotTested <62701594+WorksButNotTested@users.noreply.github.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: expend20 <36543551+expend20@users.noreply.github.com> Co-authored-by: Andrea Fioraldi <andrea.fioraldi@trellix.com> Co-authored-by: Ben Davis <ben@thebendavis.net> Co-authored-by: radl97 <radl97@users.noreply.github.com>

* step1 for tinyinst * step2: minimal executor * updated libafl * Tinyinst Update (#853) * Mac OS Autotokens (#723) * mac_tokens * more * win fix * fmt * fmt c * Use nightly fmt (#728) * Fix compilation for aarch64 qemu (#731) Typo lead to fail to compile for arm64 * Simd Fix (#729) * simd fix * fmt * Fixing readme & docs (#730) * fix * fix * add * add * fmt * 0.8.1 (#732) * New Pass Manager Arguments (#724) * new pm arguments * enable abgeana's code * Fix tui with 1 client (#734) * unbreak tui with 1 client * clippy * Add core affinity support for FreeBSD (#736) * NYX Executor (GSoC '22) (#693) * Add ccache * Update codecov.yml * Add libnyx * Fix * Add nyx build script * Fix build.sh && init executor.rs * Fix commit * Fix code * initialize `exector.rs` * refine API in `nyx_bridge.rs` * initialze `run_target` * add `test_nyxhelper` * initize `test_executor` * remove `nyx_beidge.rs` * make `test_executor` compile * Improve test * refine code * update version * fix docker * fix docker * Fix clippy * Fix build * fix build && add `set_timeout` * Fix and refine CI * fix CI * Fix CI * Add platform restrict * cargo fmt * add parallel mode * add example `nyx_libxml2_parallel` * fix fuzzer example * fix CI * add README * fix CI * fix CI * fix CI * remove unwrap and NyxResult * code format fix * add libnyx's rev * fix format * change Duration format && Fix CI * caego fmt * fix CI * fix CI * Add doc * test CI * Update test_all_fuzzers.sh * Update test_all_fuzzers.sh * Update test_all_fuzzers.sh * add cache for apt and cargo-install * Update build_and_test.yml * Update build_and_test.yml * tmp test CI * fix CI * remove debug cmd * remove test * code refine * code refine * code refine * code refine * add Makefile * fix example doc for nyx * add `NyxHelper::new_with_initial_timeout` * fix `NyxHelper::new` * fix curl parameter * code refine * add check for setup script * use afl-clang-fast in nyx * fix logic * fix makefile * fix CI * Update build_and_test.yml * Update build_and_test.yml * remove debug cmd Co-authored-by: syheliel <syheliel@gmail.com> Co-authored-by: Dominik Maier <dmnk@google.com> * Fix spelling error (#745) * OSX force_load option (#743) * Update clang.rs * fmt * Add continous JSON Logging monitor (#738) * Add simple JSON Monitor * Add documentation * Log global state * Fix formatting * Save state depending on closure outcome, have file opened all the time * Make OnDiskJSONMonitor cloneable * Switch to FnMut to allow stateful closures * Use &mut M: Monitor for the closure * Fix documentation of Rand::below (#747) * Netopenbsd build fix (#746) * core affinity netbsd implementation. * openbsd build fix * Fix autotokens doc (#751) * fix * remove wrong doc * Simplification for netbsd-specific code (#750) the cpuset api is already present in libc... * Add test case minimising stage (tmin) (#735) * add test case minimising stage * general purpose minimiser impl, with fuzzer example * reorganise, document, and other cleanup * correct python API return value * correct some docs * nit: versioning in fuzzers * ise -> ize * Implement a corpus minimiser (cmin) (#739) * initial try * correct case where cull attempts to fetch non-existent corpus entries * various on_remove, on_replace implementations * ise -> ize (consistency), use TestcaseScore instead of rolling our own * oops, feature gate * documentation! * link c++ * doc-nit: correction in opt explanation don't write documentation at 0300 * better linking * Skippable stage, generator wrapper for Grimoire (#748) * Skippable stage, generator wrapper for Grimoire * more fancy wrapper * MapFeedback: Adding support for with_name() (#752) * Adding support for with_name() * Adding with_name() function description * dragonflybsd build fix for core affinity. (#753) supporting most of linux sched api here. * CI for FreeBSD (#754) * CI for FreeBSD * rustup -y? * fixed path, switched to clippy * bsd don't source * added llvm * clippy * more yml * ? * testing ci * llvm? * llvm?? * more llvm, more tests * fixed testcase' * mem limits * more sudo * reenable all the CI * Fixes for new Clippy (#755) * New Clippy fixes for QEMU (#757) * Core affinity for FreeBSD pinning task to the wanted cpu (#756) * Do not zero-init struct in QEMU (#758) * New Clippy fixes for QEMU * no need to 0-initialize mem * clippy * Add doc for libafl_nyx (#759) Co-authored-by: syheliel <syheliel@gmail.com> * Adjust NyxExecutor trait bound to HasTargetBytes from HasBytesVec (#760) * adjust NyxExecutor trait bound to HasTargetBytes from HasBytesVec * oops actually use HasTargetBytes instead * libafl_frida: ASan hook adding Apple's memset_pattern* api. (#761) * Fix cargo doc on windows (#762) * add doc cfg * fix nostd docs * ignore CommandConfigurator doc test execution on non-unix platform * add cargo doc step pipeline on windows platform * Enable memset_patter ASan hooks for Apple on libafl_frida (#763) * Fix forkserver options (#771) * Stability improve (#773) * initial * add * fmt & fix * dbg remove * clp * clp * more * clippy * del * fix * remove unused * fix * doc * Fix doc (#780) * Add track_stability option to CalibrationStage (#781) * add * Update gramatron.rs * Update emu.rs * try * clp * Dump registers on freebsd x86_64 (#779) * Illumos support (#775) implementing core affinity too. * Reduce clang warnings for version output in libafl_cc. (#778) * Extend gramatron recursive mutator (#783) * Dump registers on NetBSD amd64 (#786) * Add support for ARMBE8 (#768) * Changes to build QEMU out-of-tree so that we don't need to clone the repo for each feature combination we build * Add be support to libafl_qemu * More config tweaks Co-authored-by: Your Name <you@example.com> * [AFLplusplus/LibAFL] dump registers on OpenBSD amd64 (PR #787) * dump registers on openbsd * write_crash implementations * Windows gdiplus (#789) * Initial steps * Harness code cleanup * don't panic on linux in order not to break the CI * formatting once again * restored cfg unix to unbreak linux build * Remove clang download from windows CI (#791) * Attempt to remove clang 12 setup * frida_gdiplus added to CI * Gdiplus comments (#792) * Attempt to remove clang 12 setup * frida_gdiplus added to CI * Redundancy note * formatting again :\ * mistake of directory name * Fix len miscalculation in grimoire string replace (#794) * Fix len miscalculation in grimoire string replace * ok Rust i was writing JS these days Co-authored-by: Andrea Fioraldi <andrea.fioraldi@trellix.com> * Fix doc typos (#796) * Fix CI (#798) * bump (#799) * Support for write_crash on netbsd (#788) * Support for bolts::cpu::read_time_counter on arm64 (#790) * Add ability to use virtual dispatch to StagesTuple (#801) * Add ability to use virtual dispatch to stagesTuple * Fix lint * Adding CPSR register for arm qemu (#800) * trying to add in observer * writing test * got up to running with instrumentation but i still need to get the map * fixing fuzzer code * adding tinyinst fuzzer * adding ffi to store all the map data into vec. * adding some new things * adding somewhat state of how i would like it should work * fixing some things * alot of false positives. * fixing before adding args * updated to use FileInput! * adding build script to pull tinyinst * fixing git issue * writing instruction to run how to run tinyinst fuzzer Co-authored-by: Dongjia Zhang <tokazerkje@outlook.com> Co-authored-by: Dominik Maier <dmnk@google.com> Co-authored-by: Phan Thanh Duy <phanthanhduypr@gmail.com> Co-authored-by: Nicholas Lang <97475577+nicklangsysdig@users.noreply.github.com> Co-authored-by: David CARLIER <devnexen@gmail.com> Co-authored-by: syheliel <45957390+syheliel@users.noreply.github.com> Co-authored-by: syheliel <syheliel@gmail.com> Co-authored-by: Aiden Hall <AidenRHall@users.noreply.github.com> Co-authored-by: Sönke <eknoes@users.noreply.github.com> Co-authored-by: Sirui Mu <msrlancern@gmail.com> Co-authored-by: Addison Crump <me@addisoncrump.info> Co-authored-by: Patrick Gersch <gersch.patrick@gmail.com> Co-authored-by: Teddy Heinen <teddy@heinen.dev> Co-authored-by: Vincent <space_white@yahoo.com> Co-authored-by: Andrea Fioraldi <andreafioraldi@gmail.com> Co-authored-by: WorksButNotTested <62701594+WorksButNotTested@users.noreply.github.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: expend20 <36543551+expend20@users.noreply.github.com> Co-authored-by: Andrea Fioraldi <andrea.fioraldi@trellix.com> Co-authored-by: Ben Davis <ben@thebendavis.net> Co-authored-by: radl97 <radl97@users.noreply.github.com> * fix * fmt * Submodule * Submodule? * Tinyinst Update V2 (#905) * updated to lastest libafl * going to replace tinyinst to more like jackalope with tinyinstrumentation * fixing clippy * keep working on cpp ffi. sad * updating litecov to tinyinst. also start making our own litecov * revert to map instead of list. not sure why its not working * making fuzzer listobserver * working with listobserver!: * cleaning up * adding cargo make run * updating cargo for tinyinst * updating readme * readme, clippy * fmt * fmt * fix * fix * docker * fix * fmt Co-authored-by: Dominik Maier <dmnk@google.com> Co-authored-by: biazo <eric.l.biazo@gmail.com> Co-authored-by: Phan Thanh Duy <phanthanhduypr@gmail.com> Co-authored-by: Nicholas Lang <97475577+nicklangsysdig@users.noreply.github.com> Co-authored-by: David CARLIER <devnexen@gmail.com> Co-authored-by: syheliel <45957390+syheliel@users.noreply.github.com> Co-authored-by: syheliel <syheliel@gmail.com> Co-authored-by: Aiden Hall <AidenRHall@users.noreply.github.com> Co-authored-by: Sönke <eknoes@users.noreply.github.com> Co-authored-by: Sirui Mu <msrlancern@gmail.com> Co-authored-by: Addison Crump <me@addisoncrump.info> Co-authored-by: Patrick Gersch <gersch.patrick@gmail.com> Co-authored-by: Teddy Heinen <teddy@heinen.dev> Co-authored-by: Vincent <space_white@yahoo.com> Co-authored-by: Andrea Fioraldi <andreafioraldi@gmail.com> Co-authored-by: WorksButNotTested <62701594+WorksButNotTested@users.noreply.github.com> Co-authored-by: Your Name <you@example.com> Co-authored-by: expend20 <36543551+expend20@users.noreply.github.com> Co-authored-by: Andrea Fioraldi <andrea.fioraldi@trellix.com> Co-authored-by: Ben Davis <ben@thebendavis.net> Co-authored-by: radl97 <radl97@users.noreply.github.com> Co-authored-by: Dominik Maier <domenukk@gmail.com>

tokatoka added 2 commits September 8, 2022 18:37

initial

adaf396

add

334a111

tokatoka marked this pull request as draft September 10, 2022 11:23

fmt & fix

e8a4d5a

tokatoka requested a review from vanhauser-thc September 10, 2022 11:55

tokatoka marked this pull request as ready for review September 10, 2022 11:55

dbg remove

e6c54db

vanhauser-thc reviewed Sep 10, 2022

View reviewed changes

libafl/src/stages/calibrate.rs Show resolved Hide resolved

clp

63089fd

vanhauser-thc reviewed Sep 10, 2022

View reviewed changes

libafl/src/stages/calibrate.rs Show resolved Hide resolved

clp

e8a642c

tsheinen reviewed Sep 10, 2022

View reviewed changes

tokatoka added 4 commits September 10, 2022 19:37

more

66a0df2

clippy

97186bd

del

a887c25

fix

1641c68

tokatoka mentioned this pull request Sep 10, 2022

Add track_stability option to CalibrationStage #776

Closed

tokatoka requested a review from domenukk September 10, 2022 20:18

remove unused

66aaf5f

tokatoka mentioned this pull request Sep 10, 2022

Illumos build fix proposal. #775

Merged

domenukk reviewed Sep 11, 2022

View reviewed changes

tokatoka added 2 commits September 12, 2022 18:03

fix

d2e929a

doc

14d7051

tokatoka merged commit d17269d into main Sep 12, 2022

tokatoka deleted the stability_improve branch September 19, 2022 13:20

khang06 pushed a commit to khang06/LibAFL that referenced this pull request Oct 11, 2022

Stability improve (AFLplusplus#773)

17b5e96

* initial * add * fmt & fix * dbg remove * clp * clp * more * clippy * del * fix * remove unused * fix * doc

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Stability improve #773

Stability improve #773

tokatoka commented Sep 10, 2022

codecov-commenter commented Sep 10, 2022 •

edited

tokatoka commented Sep 10, 2022

tsheinen Sep 10, 2022

tokatoka Sep 10, 2022

tsheinen Sep 10, 2022

WorksButNotTested Sep 10, 2022

tokatoka Sep 10, 2022

WorksButNotTested Sep 11, 2022

domenukk Sep 11, 2022

tokatoka Sep 12, 2022

domenukk Sep 11, 2022

domenukk Sep 11, 2022

domenukk Sep 11, 2022

WorksButNotTested Sep 11, 2022

domenukk Sep 11, 2022

WorksButNotTested Sep 11, 2022

tokatoka Sep 12, 2022

Stability improve #773

Stability improve #773

Conversation

tokatoka commented Sep 10, 2022

codecov-commenter commented Sep 10, 2022 • edited

Codecov Report

tokatoka commented Sep 10, 2022

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codecov-commenter commented Sep 10, 2022 •

edited