Commit Graph

12 Commits

Author SHA1 Message Date
Alexander Bulekov 0a8db249a5
qemu: limit input size to avoid timeouts (#8085)
QEMU coverage builds have been failing for a long time. Maybe this is
occuring due to the inputs being too large and timing out. Limit the
input size, in an attempt to fix this.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2022-07-26 17:01:17 +01:00
MarcoFalke 9a9a3cea1a
qemu: Simplify build script (#6468) 2021-09-17 15:02:17 +01:00
Alexander Bulekov 405818cd00
qemu: add default fuzzing options (#6411)
Close fds, and disable leak detection

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2021-09-07 12:14:11 +01:00
Alexander Bulekov 6ca77985cb
qemu: add Mauro to the auto_ccs (#6353)
Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2021-08-30 18:39:41 +01:00
Alexander Bulekov 7602748c2b
qemu: manually build glib (#5919)
* qemu: manually build glib

QEMU builds are failing [1]:
"Step #4: ERROR: glib-2.56 gthread-2.0 is required to compile QEMU"

The version of glib shipped with Ubuntu 18.04 is too old. While we wait
on [2], take some inspiration from poppler [3], and build glib ourselves.

[1] https://oss-fuzz-build-logs.storage.googleapis.com/log-2a22b4e2-d7b7-4695-a4dd-25d8d8407704.txt
[2] https://github.com/google/oss-fuzz/issues/5697
[3] a44547d8d6/projects/poppler/build.sh (L70)

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>

* qemu: add license header to build.sh

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2021-06-15 10:55:11 +01:00
Alexander Bulekov 45dca65cb9
qemu: disable honggfuzz fuzzing (#5489)
We don't do any internal tests using honggfuzz, and our code is highly
unlikely to work with honggfuzz in the current shape. All of the
true-positive bugs seem to be libfuzzer bugs.

Lets disable honggfuzz, to avoid wasting resources. We will try to get
afl++ working, before taking another look at honggfuzz.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2021-03-24 14:56:27 +00:00
Alexander Bulekov c12fc5df2c
qemu: Add necessary dependencies for 9p fuzzing (#5197)
In qemu/qemu@fff7111 we added configurations for fuzzing virtio-9p
devices. However, these fuzzers are presently stalled on oss-fuzz,
because the build image doesn't have the necessary libattr, and
libcap-ng libraries. Fix that.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2021-02-13 19:16:55 -08:00
Alexander Bulekov 7b8f46f660
qemu: add Li to the auto_ccs (#4843)
Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2020-12-14 07:02:28 -08:00
Oliver Chang 84faf55729
Populate a bunch of main_repo values. (#4815)
Also enforce this for future integrations.
2020-12-10 09:49:27 +11:00
Alexander Bulekov 4bf509a916
[qemu] add ninja to the build image (#4557)
As of QEMU's 09e93326e4 ("build: replace ninjatool with ninja"), QEMU
now requires ninja instead of ninjatool for builds.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2020-10-21 10:43:56 -07:00
alex da7bb14b96
QEMU: Add fuzz target build scripts (#3935)
* QEMU: Add fuzz target build scripts

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>

* QEMU: Disable AFL and MSan in project.yaml

We can't link against OSS-Fuzz' afl driver as it is not compiled with
-fPIC. Additionally, it appears that we trigger some false-positives
with MemorySanitizer, since we don't instrument all of the libraries.
Disable AFL and MemorySanitizer, for now.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
2020-06-21 16:38:41 -07:00
alex 75e1d2d7ab
Add QEMU Project Description (#3526) 2020-03-20 13:27:56 -07:00