Add a library suitable for libfuzzer with a small unit test runner shell

Add a library suitable for libfuzzer with a small unit test runner shell

BUG=chromium:577261
R=machenbach@chromium.org,jarin@chromium.org
LOG=n

Committed: https://crrev.com/cf0435881c66d8a41b9cd923d9e885065c721ebf
Cr-Commit-Position: refs/heads/master@{#33508}

[jochen (gone - plz use gerrit), 2016-01-20 10:29:11]

Michael, can you help me figuring out how to hook this up to run-test.py?

I'd like to eventually have several ${foo}_fuzzer binaries, that all take inputs
from test/fuzzer/${foo}

I couldn't get a single test/fuzzer/testcfg.py that would run several binaries,
nor does run-tests.py seem to support test suites in subdirectories for
test/fuzzer/?

[Michael Achenbach, 2016-01-20 10:47:17]

https://codereview.chromium.org/1604203002/diff/1/build/all.gyp
File build/all.gyp (right):

https://codereview.chromium.org/1604203002/diff/1/build/all.gyp#newcode28
build/all.gyp:28: '../test/fuzzer/fuzzer.gyp:*',
Guess that's redundant, since it's covered unconditionally above.

https://codereview.chromium.org/1604203002/diff/1/tools/run-tests.py
File tools/run-tests.py (right):

https://codereview.chromium.org/1604203002/diff/1/tools/run-tests.py#newcode71
tools/run-tests.py:71: "parser_fuzzer",
This list only supports toplevel entries in v8/test/*. E.g. here: "fuzzer". A
nesting feature does not yet exist.

[Jarin, 2016-01-22 13:50:18]

The .(cc|h) part lgtm.

https://codereview.chromium.org/1604203002/diff/20001/test/fuzzer/fuzzer.cc
File test/fuzzer/fuzzer.cc (right):

https://codereview.chromium.org/1604203002/diff/20001/test/fuzzer/fuzzer.cc#n...
test/fuzzer/fuzzer.cc:43: fread(data, 1, size, input);
As gcc suggests, check that the return value is the same as 'size'.

[jochen (gone - plz use gerrit), 2016-01-25 14:45:31]

ptal

[Michael Achenbach, 2016-01-25 15:05:53]

Got a feeling that the tests are not executed on any bot...

https://codereview.chromium.org/1604203002/diff/40001/test/fuzzer/testcfg.py
File test/fuzzer/testcfg.py (right):

https://codereview.chromium.org/1604203002/diff/40001/test/fuzzer/testcfg.py#...
test/fuzzer/testcfg.py:11: class FuzzerTestSuite(testsuite.TestSuite):
Do our testing variants make sense for these tests (i.e. nocrankshaft, turbo,
normal, etc)? If not, you might want to add something like this:
https://code.google.com/p/chromium/codesearch#chromium/src/v8/tools/testrunne...

https://codereview.chromium.org/1604203002/diff/40001/tools/run-tests.py
File tools/run-tests.py (right):

https://codereview.chromium.org/1604203002/diff/40001/tools/run-tests.py#newc...
tools/run-tests.py:71: "fuzzer",
This is currently ordered by expected slowest -> fastest, so that
multiprocessing gets most throughput. Will "fastest" hold for these tests? If in
doubt, put it somewhere in the middle, i.e. after cctest.

[Michael Achenbach, 2016-01-25 15:25:13]

https://codereview.chromium.org/1604203002/diff/40001/tools/run-tests.py
File tools/run-tests.py (right):

https://codereview.chromium.org/1604203002/diff/40001/tools/run-tests.py#newc...
tools/run-tests.py:63: "bot_default": [
This should stay in sync with:
https://code.google.com/p/chromium/codesearch#chromium/src/v8/test/bot_defaul...
and
https://code.google.com/p/chromium/codesearch#chromium/src/v8/test/default.is...

+ the two gyp files there.

This might deserve a comment to not fall into the same trap :/

[jochen (gone - plz use gerrit), 2016-01-26 07:54:20]

ptal

I now see the test in the swarming results

https://codereview.chromium.org/1604203002/diff/40001/test/fuzzer/testcfg.py
File test/fuzzer/testcfg.py (right):

https://codereview.chromium.org/1604203002/diff/40001/test/fuzzer/testcfg.py#...
test/fuzzer/testcfg.py:11: class FuzzerTestSuite(testsuite.TestSuite):
On 2016/01/25 at 15:05:53, Michael Achenbach wrote:
> Do our testing variants make sense for these tests (i.e. nocrankshaft, turbo,
normal, etc)? If not, you might want to add something like this:
>
https://code.google.com/p/chromium/codesearch#chromium/src/v8/tools/testrunne...

some systems might behave differently based on those flags

[Michael Achenbach, 2016-01-26 08:05:52]

lgtm with comment:

Could you also add fuzzer.gyp here:
https://code.google.com/p/chromium/codesearch#chromium/src/v8/test/bot_defaul...
https://code.google.com/p/chromium/codesearch#chromium/src/v8/test/default.gyp

Usually we build the ninja All target, but e.g. auto-bisect rebuilds only the
target with the failing test, e.g. only bot_default.

[jochen (gone - plz use gerrit), 2016-01-26 10:13:15]

The CQ bit was checked by jochen@chromium.org

[jochen (gone - plz use gerrit), 2016-01-26 10:13:15]

The patchset sent to the CQ was uploaded after l-g-t-m from jarin@chromium.org,
machenbach@chromium.org
Link to the patchset: https://codereview.chromium.org/1604203002/#ps100001
(title: "updates")

[commit-bot: I haz the power, 2016-01-26 10:13:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1604203002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1604203002/100001

[Michael Achenbach, 2016-01-26 10:14:42]

lgtm

[commit-bot: I haz the power, 2016-01-26 10:38:43]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2016-01-26 10:39:10]

Description was changed from

==========
Add a library suitable for libfuzzer with a small unit test runner shell

BUG=chromium:577261
R=machenbach@chromium.org,jarin@chromium.org
LOG=n
==========

to

==========
Add a library suitable for libfuzzer with a small unit test runner shell

BUG=chromium:577261
R=machenbach@chromium.org,jarin@chromium.org
LOG=n

Committed: https://crrev.com/cf0435881c66d8a41b9cd923d9e885065c721ebf
Cr-Commit-Position: refs/heads/master@{#33508}
==========

[commit-bot: I haz the power, 2016-01-26 10:39:11]

Patchset 6 (id:??) landed as
https://crrev.com/cf0435881c66d8a41b9cd923d9e885065c721ebf
Cr-Commit-Position: refs/heads/master@{#33508}