Message-ID: <CAEwNFnDJzXZoqsSo3p5XVZk3ux+6TfjRzWJYxgMuiUVTJYiRYQ@mail.gmail.com>
Date:	Thu, 11 Aug 2011 08:50:43 +0900
From:	Minchan Kim <minchan.kim@...il.com>
To:	Mel Gorman <mgorman@...e.de>
Cc:	linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: MMTests 0.01

Hi Mel,

At last, you release the great test scripts.
Awesome! I really welcome this!

We have needed standard test suite to discuss and test more easily.
And it can help to find regression through periodic test.

Of course, it would be good if LTP or autotest merge this tests.
But I think it's not bad that we maintains it with separate test for
heading goal of mm-specific standard test suite. :) For it, at least,
we need public git tree.

Anyway, Thanks for sharing your valuable knowhow, Mel.

On Thu, Aug 4, 2011 at 11:38 PM, Mel Gorman <mgorman@...e.de> wrote:
> At LSF/MM at some point a request was made that a series of tests be
> identified that were of interest to MM developers and that could be
> used for testing the Linux memory management subsystem. At the time,
> I was occasionally posting tarballs of whatever scripts I happened to
> be using at the time but they were not generally usable and tended to
> be specific to a set of patches. I promised I would produce something
> usable by others but never got around to it. Over the last four months,
> I needed a better framework when testing against both distribution
> kernels and mainline so without further ado
>
> http://www.csn.ul.ie/~mel/projects/mmtests/
> http://www.csn.ul.ie/~mel/projects/mmtests/mmtests-0.01-mmtests-0.01.tar.gz
>
> I am not claiming that this is comprehensive in any way but it is
> almost always what I start with when testing patch sets. In preparation
> for identifying problems with backports, I also ran a series of tests
> against mainline kernels over the course of two months when machines
> were otherwise idle. I have not actually had a chance to go through
> all the results and identify each problem but I needed to have the
> raw data available for my own reference so might as well share.
>
> http://www.csn.ul.ie/~mel/projects/mmtests/results/SLES11sp1/
> http://www.csn.ul.ie/~mel/projects/mmtests/results/openSUSE11.4/
>
> The directories refer to the distribution used but not the
> kernel which is downloaded from kernel.org. Directory structure is
> distro/config/machine/comparison.html. For example a set of benchmarks
> used for evaluating the page and slab allocators on a test machine
> called "hydra" is located at
>
> http://www.csn.ul.ie/~mel/projects/mmtests/results/SLES11sp1/global-dhp__pagealloc-performance/hydra/comparison.html
>
> I know the report structure looks crude but I was not interested
> in making them pretty. Due to the fact that some of the scripts
> are extremely old, the quality and coding styles vary considerably.
> This may get cleaned up over time but in the meantime, try and keep
> the contents of your stomach down if you are reading the scripts.
>
> The documentation is not great and so some of the capabilities such
> as being able to reconfigure swap for a benchmark is not mentioned.
> For my own series, I'll relase the mmtests tarball I used if asked.
> If someone wants to use the tarball for their own testing but cannot
> configure it, complain on the linux-mm list and if I can, I'll offer
> suggestions.
>
> ==== MMTests README ====
>
> MMTests is a configurable test suite that runs a number of common
> workloads of interest to MM developers. Ideally this would have been
> to integrated with LTP, xfstests or Phoronix Test or implemented
> with autotest.  Unfortunately, large portions of these tests are
> cobbled together over a number of years with varying degrees of
> quality before decent test frameworks were common.  The refactoring
> effort to integrate with another framework is significant.
>
> Organisation
> ============
>
> The top-level directory has a single driver script called
> run-mmtests.sh which reads a config file that describes how the
> benchmarks should be run, configures the system and runs the requested
> tests. config also has some per-test configuration items that can be
> set depending on the test. The driver script takes the name of the
> test as a parameter. Generally, this would be a symbolic name naming
> the kernel being tested.
>
> Each test is driven by a run-single-test.sh script which reads
> the relevant driver-TESTNAME.sh script. High level items such as
> profiling are configured from the top-level script while the driver
> scripts typically convert the config parameters into switches for a
> "shellpack". A shellpack is a pair of benchmark and install scripts
> that are all stored in shellpacks/ .
>
> Monitors can be optionally configured. A full list is in monitors/
> . Care should be taken with monitors as there is a possibility that
> they introduce overhead of their own.  Hence, for some performance
> sensitive tests it is preferable to have no monitoring.
>
> Many of the tests download external benchmarks. An attempt will be
> made to download from a mirror . To get an idea where the mirror
> should be located, grep for MIRROR_LOCATION= in shellpacks/.
>
> A basic invocation of the suite is
>
> <pre>
> $ cp config-global-dhp__pagealloc-performance config
> $ ./run-mmtests.sh --no-monitor 3.0-nomonitor
> $ ./run-mmtests.sh --run-monitor 3.0-runmonitor
> </pre>
>
> Configuration
> =============
>
> The config file used is always called "config". A number of other
> sample configuration files are provided that have a given theme. Some
> important points of variability are;
>
> MMTESTS is a list of what tests will be run
>
> WEBROOT is the location where a number of tarballs are mirrored. For example,
>        kernbench tries to download
>        $WEBROOT/kernbench/linux-2.6.30.tar.gz . If this is not available,
>        it is downloaded from the internet. This can add delays in testing
>        and consumes bandwidth so is worth configuring.
>
> LINUX_GIT is the location of a git repo of the kernel. At the moment it's only
>        used during report generation
>
> SKIP_*PROFILE
>        These parameters determine what profiling runs are done. Even with
>        profiling enabled, a non-profile run can be used to ensure that
>        the profile and non-profile runs are comparable.
>
> SWAP_CONFIGURATION
> SWAP_PARTITIONS
> SWAP_SWAPFILE_SIZEMB
>        It's possible to use a different swap configuration than what is
>        provided by default.
>
> TESTDISK_RAID_PARTITIONS
> TESTDISK_RAID_DEVICE
> TESTDISK_RAID_OFFSET
> TESTDISK_RAID_SIZE
> TESTDISK_RAID_TYPE
>        If the target machine has partitions suitable for configuring RAID,
>        they can be specified here. This RAID partition is then used for
>        all the tests
>
> TESTDISK_PARTITION
>        Use this partition for all tests
>
> TESTDISK_FILESYSTEM
> TESTDISK_MKFS_PARAM
> TESTDISK_MOUNT_ARGS
>        The filesystem, mkfs parameters and mount arguments for the test
>        partitions
>
> Available tests
> ===============
>
> Note the ones that are marked untested. These have been ported from other
> test suites but no guarantee they actually work correctly here. If you want
> to run these tests and run into a problem, report a bug.
>
> kernbench
>        Builds a kernel 5 times recording the time taken to completion.
>        An average time is stored. This is sensitive to the overall
>        performance of the system as it hits a number of subsystems.
>
> multibuild
>        Similar to kernbench except it runs a number of kernel compiles
>        in parallel. Can be useful for stressing the system and seeing
>        how well it deals with simple fork-based parallelism.
>
> aim9
>        Runs a short version of aim9 by default. Each test runs for 60
>        seconds. This is a micro-benchmark of a number of VM operations. It's
>        sensitive to changes in the allocator paths for example.
>
> vmr-stream
>        Runs the STREAM benchmark a number of times for varying sizes. An
>        average is recorded. This can be used to measure approximate memory
>        throughput or the average cost of a number of basic operations. It is
>        sensitive to cache layout used for page faults.
>
> vmr-cacheeffects (untested)
>        Performs linear and random walks on nodes of different sizes stored in
>        a large amount of memory. Sensitive to cache footprint and layout.
>
> vmr-createdelete (untested)
>        A micro-benchmark that measures the time taken to create and delete
>        file or anonymous mappings of increasing sizes. Sensitive to changes
>        in the page fault path performance.
>
> iozone
>        A basic filesystem benchmark.
>
> fsmark
>        This tests write workloads varying the number of files and directory
>        depth.
>
> hackbench-*
>        Hackbench is generally a scheduler benchmark but is also sensitive to
>        overhead in the allocators and to a lesser extent the fault paths.
>        Can be run for either sockets or pipes.
>
> largecopy
>        This is a simple single-threaded benchmark that downloads a large
>        tar file, expands it a number of times, creates a new tar and
>        expands it again. Each operation is timed and is aimed at shaking
>        out stall-related bugs when copying large amounts of data
>
> largedd
>        Similar to largecopy except it uses dd instead of cp.
>
> libreofficebuild
>        This downloads and builds libreoffice. It is a more aggressive
>        compile-orientated test. This is a very download-intensive
>        benchmark and was only created as a reproduction case for
>        a bug.
>
> nas-*
>        The NAS Parallel Benchmarks for the serial and openmp versions of
>        the test.
>
> netperf-*
>        Runs the netperf benchmark for *_STREAM on the local machine.
>        Sensitive to cache usage and allocator costs. To test for cache line
>        bouncing, the test can be configured to bind to certain processors.
>
> postmark
>        Run the postmark benchmark. Optionally a program can be run in
>        the background that consumes anonymous memory. The background
>        program is vary rarely needed except when trying to identify
>        desktop stalls during heavy IO.
>
> speccpu (untested)
>        SPECcpu, what else can be said. A restriction is that you must have
>        a mirrored copy of the tarball as it is not publicly available.
>
> specjvm (untested)
>        SPECjvm. Same story as speccpu
>
> specomp (untested)
>        SPEComp. Same story as speccpu
>
> sysbench
>        Runs the complex workload for sysbench backed by postgres. Running
>        this test requires a significant build environment on the test
>        machine. It can run either read-only or read/write tests.
>
> simple-writeback
>        This is a simple writeback test based on dd. It's meant to be
>        easy to understand and quick to run. Useful for measuring page
>        writeback changes.
>
> ltp (untested)
>        The LTP benchmark. What it is testing depends on exactly which of the
>        suite is configured to run.
>
> ltp-pounder (untested)
>        ltp pounder is a non-default test that exists in LTP. It's used by
>        IBM for hardware certification to hammer a machine for a configured
>        number of hours. Typically, they expect it to run for 72 hours
>        without major errors.  Useful for testing general VM stability in
>        high-pressure low-memory situations.
>
> stress-highalloc
>        This test requires that the system not have too much memory and
>        that systemtap is available. Typically, it's tested with 3GB of
>        RAM. It builds a number of kernels in parallel such that total
>        memory usage is 1.5 times physical memory. When this is running
>        for 5 minutes, it tries to allocate a large percentage of memory
>        (e.g. 95%) as huge pages recording the latency of each operation as it
>        goes. It does this twice. It then cancels the kernel compiles, cleans
>        the system and tries to allocate huge pages at rest again. It's a
>        basic test for fragmentation avoidance and the performance of huge
>        page allocation.
>
> xfstests (untested)
>        This is still at prototype level and aimed at running testcase 180
>        initially to reproduce some figures provided by the filesystems people.
>
> Reporting
> =========
>
> For reporting, there is a basic compare-kernels.sh script. It must be updated
> with a list of kernels you want to compare and in what order. It generates a
> table for each test, operation and kernel showing the relative performance
> of each. The test reporting scripts are in subreports/. compare-kernel.sh
> should be run from the path storing the test logs. By default this is
> work/log. If you are automating tests from an external source, work/log is
> what you should be capturing after a set of tests complete.
>
> If monitors are configured such as ftrace, there are additional
> processing scripts. They can be activated by setting FTRACE_ANALYSERS in
> compare-kernels.sh. A basic post-process script is mmtests-duration which
> simply reports how long an individual test took and what its CPU usage was.
>
> There are a limited number of graphing scripts included in report/
>
> TODO
> ====
>
> o Add option to test on filesystem loopback device stored on tmpfs
> o Add volanomark
> o Create config-* set suitable for testing scheduler to isolate situations
>  where the scheduler was the main cause of a regression
>
> --
> Mel Gorman
> SUSE Labs
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@...ck.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
> Don't email: <a href=mailto:"dont@...ck.org"> email@...ck.org </a>
>



-- 
Kind regards,
Minchan Kim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists