* [infra] Add the build configuration and other pieces for clang code coverage.
* Address the initial feedback, still a draft with some pieces missing.
* Trying to test things locally.
* Work in progress, now able to test the build process.
* Rebase and fix a comment.
* Figure out how to upload report, but still have issues with GCS access.
* Remove unnecessary debug prints.
* WIP: adding targets list reading + minor clean up.
* WIP
* WIP: now it works end to end but is very dirty and has limitations.
* WIP: use rsync -r -d to overwrite older reports (if any)
* WIP: trying to use a helper script for bulk corpus download.
* WIP: download corpus in batches passed via args.
* WIP: testing corpus downloading in batches.
* WIP: woohoo, now it works!
* Some cleanup for a high level review.
* Address review comments and add FAQ entry.
* Update FAQ.md as per comments by Oliver.
Turns out we need to be in out/proc/self/cwd, not out/src/proc/self/cwd.
In #1721, I was picking up stale artifacts from manual CLI rsync.
Signed-off-by: Harvey Tuch <htuch@google.com>
* [infra] profile command: do not fail hard if some fuzz targets have no corpus.
* Change listing to corpus_listing to be more explicit.
* Address review comment from Abhishek
This will make it easier for us to control individual test targets and
exclude trivials/bad fuzzers from the Envoy GH repo.
Signed-off-by: Harvey Tuch <htuch@google.com>
* [infra] Add script for downloading a batch of corpus backups (for #1547).
* Add missing wget dependency and replace a tab with spaces.
* Convert more tabs to spaces.
* Adds gcrypt to elliptic curve diff fuzzer
And adds corpus to the project
* A way to solve conflict between openssl and gcrypt
* Code review : no longer use of curl nor sed
* better seed corpus for fuzz_key
* new target: fuzz_variant_text
* new target: fuzz_dbus_message
* get ninja from pip
* remove target: fuzz_markup
* new target: fuzz_variant_binary
* Add Firestore project
* Firestore project docker image and build script
* Firestore project docker image and build script
* Clone Firebase from Git and build with Fuzzing
* Add a collaborator and comments for cmake version
* Moving fuzzing target to test folder
When a fuzzer is seriously broken (which happens occasionally during debug),
it's unlikely to start properly let alone provide some meaningful
output. In this case, it seems reasonable to skip some checks and prevent bash
from encountering the following syntax error:
```
/usr/local/bin/bad_build_check: line 68: ((: < 100 : syntax error: operand expected (error token is "< 100 ")
```
* Generic listing of fuzz targets for project gnupg
So that we will no longer need pull requests on oss-fuzz
when one new fuzz target is added in gnupg code
There is one new fuzz target : fuzz_list
* Updating libgpg-error to version 1.32
To get the ball rolling more quickly, the download_refs fuzzer for
libgit2 was first implemented directly in oss-fuzz. But as we want to
ensure that we're not breaking fuzzers and thus at least build them as
part of our CI, the goal has been to upstream them into libgit2.
This commit removes the download_refs fuzzer and its corpus in favor of
using the upstreamed fuzzer. Furthermore, the build script is
generalized to automatically pick up new fuzzers named according to a
certain pattern, as we already added a second packfile fuzzer.
Max Moroz