[Piglit] [PATCH] deqp: Add test profile for external dEQP-GLES3 tests
Mike Mason
michael.w.mason at intel.com
Wed Dec 10 09:03:33 PST 2014
Hi Chad,
See comments below...
On Sunday, December 07, 2014 08:54:06 PM Chad Versace wrote:
> Google opensourced the drawElements Quality Product Testsuite (dEQP) as
> part of the Android Lollipop release. It's git repo lives in the Android
> tree at [https://android.googlesource.com/platform/external/deqp/].
>
> This patch adds a new module, deqp_gles3.py, that runs the dEQP-GLES3
> tests. It queries the 'deqp-gles3' executable at runtime for the list
> of testcases. The module attempts queries the 'deqp-gles3' executable
> only if the environment variable PIGLIT_DEQP_GLES3_EXE or the
> piglit.conf option deqp-gles3.exe is set.
>
> Why do we need to use Pigit as the testrunner for dEQP? (After all, dEQP has
> its own testrunner). Because dEQP runs all tests in a single process. If test
> 17530 of 55409 crashes, then the dEQP testrunner crashes and the remaining
> tests never run. Piglit doesn't suffer from that problem, because it runs each
> test as a separate process.
>
> Tested on Intel Ivybridge by running the command
> piglit run -t dEQP-GLES3/info/vendor \
> tests/deqp_gles3.py tests/quick.py results
> with and without PIGLIT_DEQP_GLES3_EXE set and with and without
> piglit.conf:deqp-gles.exe set.
>
> Cc: Dylan Baker <dylanx.c.baker at intel.com>
> Cc: Daniel Kurtz <djkurtz at chromium.org>
> Cc: Ilja H. Friedel <ihf at chromium.org>
> Signed-off-by: Chad Versace <chad.versace at linux.intel.com>
> ---
>
> You can find this patch on top of my 'deqp-gles3' branch.
>
> git://github.com/chadversary/piglit refs/heads/deqp-gles3
>
> Daniel and Ilja, I CC'd you just so you would be aware of the dEQP/Piglit work
> Intel is doing. It would be nice to see it go into Chrome OS enventually.
>
>
> piglit.conf.example | 7 +++
> tests/deqp_gles3.py | 137 ++++++++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 144 insertions(+)
> create mode 100644 tests/deqp_gles3.py
>
> diff --git a/piglit.conf.example b/piglit.conf.example
> index a864c01..0ac0f4f 100644
> --- a/piglit.conf.example
> +++ b/piglit.conf.example
> @@ -33,6 +33,13 @@ bindir=/home/usr/oclconform
> testA
> testB
>
> +;[deqp-gles3]
> +;
> +; Path to the deqp-gles3 executable. You can also set this with the environment
> +; variable PIGLIT_DEQP_GLES3_EXE. Piglit will run the dEQP-GLES3 tests if and
> +; only if this option is set.
> +;exe=/home/knuth/deqp/modules/gles3/deqp-gles3
> +
> ; Section for specific oclconform test. One of these sections is required for
> ; each test list in the oclconform section and must be called:
> ; oclconform-$testname
> diff --git a/tests/deqp_gles3.py b/tests/deqp_gles3.py
> new file mode 100644
> index 0000000..422d1fa
> --- /dev/null
> +++ b/tests/deqp_gles3.py
> @@ -0,0 +1,137 @@
> +# Copyright 2014 Intel Corporation
> +#
> +# Permission is hereby granted, free of charge, to any person obtaining a copy
> +# of this software and associated documentation files (the "Software"), to deal
> +# in the Software without restriction, including without limitation the rights
> +# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
> +# copies of the Software, and to permit persons to whom the Software is
> +# furnished to do so, subject to the following conditions:
> +#
> +# The above copyright notice and this permission notice shall be included in
> +# all copies or substantial portions of the Software.
> +#
> +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
> +# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> +# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
> +# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
> +# SOFTWARE.
> +
> +import ConfigParser
> +import os
> +import shutil
> +import subprocess
> +import sys
> +import tempfile
> +
> +import framework.profile
> +
> +
> +__all__ = ['profile']
> +
> +
> +_deqp_gles3_exe = None
> +
> +
> +def get_option(env_varname, config_option):
> + """Query the given environment variable and then piglit.conf for the
> + option. Return None if the option is unset.
> + """
> +
> + opt = os.environ.get(env_varname, None)
> + if opt is not None:
> + return opt
> +
> + try:
> + opt = framework.core.PIGLIT_CONFIG.get(config_option[0],
> + config_option[1])
> + except ConfigParser.NoSectionError, ConfigParser.NoOptionError:
> + pass
> +
> + return opt
> +
> +
> +def get_deqp_gles3_exe():
> + """Get path to the deqp-gles3 executable. Idempotent."""
> + global _deqp_gles3_exe
> +
> + if _deqp_gles3_exe is None:
> + _deqp_gles3_exe = get_option('PIGLIT_DEQP_GLES3_EXE',
> + ('deqp-gles3', 'exe'))
> +
> + return _deqp_gles3_exe
> +
> +
> +def gen_caselist_txt():
> + """Generate dEQP-GLES3-cases.txt and return its path."""
> + # dEQP is stupid (2014-12-07):
> + # 1. To generate the caselist file, dEQP requires that the process's
> + # current directory must be that same as that of the executable.
> + # Otherwise, it fails to find its data files.
> + # 2. dEQP creates the caselist file in the process's current directory
> + # and provides no option to change its location.
> + # 3. dEQP creates a GL context when generating the caselist. Therefore,
> + # the caselist must be generated on the test target rather than the
> + # build host. In other words, when the build host and test target
> + # differ then we cannot pre-generate the caselist on the build host:
> + # we must *dynamically* generate it during the testrun.
> + basedir = os.path.dirname(get_deqp_gles3_exe())
> + caselist_path = os.path.join(basedir, 'dEQP-GLES3-cases.txt')
> + subprocess.check_call(
> + [get_deqp_gles3_exe(), '--deqp-runmode=txt-caselist'],
> + cwd=basedir)
> + assert(os.path.exists(caselist_path))
> + return caselist_path
> +
> +
> +def iter_deqp_test_cases():
> + """Iterate over original dEQP GLES3 testcase names."""
> + caselist_path = gen_caselist_txt()
> + with open(caselist_path) as caselist_file:
> + for i, line in enumerate(caselist_file):
> + if line.startswith('GROUP:'):
> + continue
> + elif line.startswith('TEST:'):
> + yield line[len('TEST:'):].strip()
> + else:
> + raise Exception('{0}:{1}: ill-formed line'.format(caselist_txt, i))
^^^^^^^^^^^^
caselist_txt doesn't exist. Did you mean to use line or caselist_path?
> +
> +
> +class DEQPTest(framework.profile.Test):
> +
> + def __init__(self, case_name):
> + command = [
> + get_deqp_gles3_exe(),
> + '--deqp-case=' + case_name,
> + '--deqp-visibility=hidden',
> + ]
> + framework.profile.Test.__init__(self, command)
> +
> + # dEQP's working directory must be the same as that of the executable,
> + # otherwise it cannot find its data files (2014-12-07).
> + self.cwd = os.path.dirname(get_deqp_gles3_exe())
> +
> + def interpret_result(self):
> + if self.result['returncode'] == 0:
> + self.result['result'] = 'pass'
> + else:
> + self.result['result'] = 'fail'
The script interprets deqp results is incorrectly. If the binary
runs and exits cleanly it returns 0 regardless of the actual test result.
I verified this by testing the return code of a failed test.
The real result is in self.result[‘out’].
Here’s a version of interpret_result that takes that into account:
def interpret_result(self):
result_map = {"Pass" : "pass",
"Fail" : "fail",
"QualityWarning" : "fail",
"InternalError" : "fail",
"Crash" : "fail",
"NotSupported" : "skip"}
if self.result['returncode'] != 0:
self.result['result'] = 'fail'
return
for line in self.result['out'].split('\n'):
line = line.lstrip()
for result in result_map:
if line.startswith(result):
self.result['result'] = result_map[result]
return
> +
> +def add_tests():
> + global profile
> +
> + if get_deqp_gles3_exe() is None:
> + return
> +
> + for deqp_testname in iter_deqp_test_cases():
> + # dEQP uses '.' as the testgroup separator.
> + piglit_testname = deqp_testname.replace('.', '/')
> +
> + test = DEQPTest(deqp_testname)
> + profile.test_list[piglit_testname] = test
> +
> +
> +profile = framework.profile.TestProfile()
> +add_tests()
> --
> 1.9.3
>
> _______________________________________________
> Piglit mailing list
> Piglit at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/piglit
>
More information about the Piglit
mailing list