1
0
Fork 0
mirror of https://github.com/VSadov/Satori.git synced 2025-06-11 18:20:26 +09:00
Commit graph

21 commits

Author SHA1 Message Date
Brian Robbins
dee362eb25 Update Performance Test Execution Instructions (dotnet/corefx#27475)
Commit migrated from bbfc8856fd
2018-02-26 12:11:46 -08:00
Dan Moseley
549222c3d2 Update performance-tests.md
Commit migrated from 1678ac268e
2018-02-03 20:01:06 -08:00
Steve Franceschelli
c6e139b15a Add note to build -release before running perf tests.
Commit migrated from bc378e2b9c
2017-05-04 20:16:42 -04:00
stephenmichaelf
d6d61fdf7b Update performance-tests.md (dotnet/corefx#19158)
Commit migrated from 0c379b25ae
2017-05-01 17:23:58 -07:00
Bill Wert
ac379f0236 Update guidance to include special types
I noticed a handful of tests substituting int for a value type, which
does not always do what you might want. In general test authors should
consider if the codepath being tested has optimizations around built-in
data types or get faster if types implement certain interfaces.


Commit migrated from fa8a38b312
2017-04-30 20:29:05 -07:00
Eric Mellino
c4ae964bd9 Fix typo and reword sentence.
Commit migrated from 4d3384bb1b
2017-04-24 13:11:54 -07:00
Eric Mellino
f867924f22 Add extra performance test guidance, based on PR feedback
Commit migrated from 2124d36789
2017-04-24 13:10:34 -07:00
Eric Mellino
caace37c4d Clarify a point in perf-testing.md.
Commit migrated from 0194225ce4
2017-04-21 11:39:47 -07:00
Eric Mellino
bd704b1ef2 Update code example and add extra bullet points.
Commit migrated from e0c7181e3c
2017-04-21 10:39:32 -07:00
Eric Mellino
8e7ac4c35e Fix extra brace in code example.
Commit migrated from 93c476c4be
2017-04-20 14:53:12 -07:00
Eric Mellino
0f9f347b94 Update performance-tests.md with benchmark-writing guidance
Adds a short list of principles to follow when writing performance tests.

Commit migrated from a63a31c82c
2017-04-20 14:50:57 -07:00
Ahson Ahmed Khan
295ddf21f3 Fixing typo in docs - /p:OSGroup => /p:TargetOS (dotnet/corefx#17319)
Commit migrated from 850b0f9400
2017-03-20 20:40:38 -07:00
James Ko
77faa41ae2 Docs: Add /p:OSGroup to performance testing parameters
Commit migrated from 79bd427770
2016-08-29 18:09:58 -04:00
Stephen A. Imhoff
6fb98b049f Correct parameter name.
Commit migrated from 363c45c540
2016-08-24 18:14:02 -07:00
Eric St. John
5a444416fd Update tests to use NETCoreApp1.0 moniker
Requires https://github.com/dotnet/buildtools/pull/751

This makes all the tests resolve using NETCoreApp1.0 instead of
DNXCore50.  I've removed the netstandard1.6 import since that is no
longer needed.  I've also removed the workarounds from tests that were
targeting netcoreapp1.0 on their own.  For restore compat I've added an
DNXCore50 import to every test project.  These can be removed once
test-runtime packages are updated to target NETCoreApp1.0.

I've updated all the test scripts which were assuming DNXCore50 as well
as docs which referred to this path.


Commit migrated from 7ba4d296e3
2016-05-24 21:49:17 -07:00
Matt Ellis
56273574df Consume v3 versions of the new feeds
- We have moved our feeds to our enterprise myget
  account (dotnet.myget.org). Start pulling from there and use the v3
  feeds since we are restoring with tools that understand them.

- Remove some no longer needed NuGet commands and sources


Commit migrated from 6b2396f986
2016-02-05 15:54:05 -08:00
Ian Hays
851c786922 Perf test targets were buggy and required specific setup to not fail. I've hardened them so that the tests can be more easily built without also running them. I also removed the v5.0 libraries requirement. This commit also allows the perf tests to be run as regular unit tests.
Commit migrated from fa1635c07d
2015-11-04 07:58:23 -08:00
Ian Hays
1a12a1d25d Update and addition to Performance tests
- Made the existing perf tests for Linq follow the new syntax and use the new runner
- Removed all references to v5.0 libraries being required as they no longer are
- Added new tests using usage data from aspnet/benchmarks
- Modified some existing tests to be more easily comparable using analysis tools that look at inlinedata/memberdata (e.g. System.IO.Compression)
- Updated the lock files to use the latest xunit-performance package
- Modified test structure and innerIteration loops to produce more consistently repeatable results
- Updated docs with change to csproj requirements


Commit migrated from 2ad9a6ffa7
2015-10-13 14:28:01 -07:00
Ian Hays
1a02314f02 PR feedback
Commit migrated from 6ca45dedc4
2015-10-02 11:51:15 -07:00
Ian Hays
de822c50f0 Updated perf tests and documentation
- Updated getting started perf docs
- Removed caching of collections for tests; my results found the caching of negligible benefit
- Changed from MemberData to InlineData where possible to simplify matching method signatures for the perf analysis tools.
- Fixed the namespace for FileSystem tests to be the same as the other FS tests


Commit migrated from f551cdfe7b
2015-10-02 11:37:12 -07:00
Ian Hays
01468ad85f Added performance microbenchmarks
- Added "Performance" folders containing perf xunit tests to a number of highly-used libraries (determined by Usage data)
- Two sets of usage data from store apps; I added a test for every function used in more than 5% of the covered assemblies in both data sets.
- This is a first-round pass of the most highly used libraries and does not include: abstract classes, interfaces, or Attributes
- Created a static utility class for commonly used functions across perf tests, stored in src\Common\tests\Performance.
- These tests will run using ```msbuild /t:BuildAndTest /p:performance=true``` from the projects test directory
- Added conditional properties that check if v5.0 is installed on the user's computer when running perf tests. If it isn't, a detailed error is thrown with a link to a performance-test github page that I've also added.


Commit migrated from 11ea534258
2015-09-24 10:01:40 -07:00