Merged
Conversation
This reverts commit 4223cc2.
Co-authored-by: Rainer Sigwald <raines@microsoft.com>
Co-authored-by: Rainer Sigwald <raines@microsoft.com>
Works around win32 limitation--a group will report only up to 32 processors on a 32-bit process.
MSBuild normalizes slashes such that a solution file can use backslashes in paths on Linux or iOS. This adds that functionality for solution filter files as well. Fixes dotnet#5715.
* Added .swr file. Try catching the program to ensure it doesn't fail and thus prevent MSBuild from installing * Moved exe.swr to its own directory. Otherwise both .swr files are passed to arcade as a semicolon delimited list, which arcade can't deal with * Sign UnGAC.exe
Fixes dotnet#5736 by keeping a prototype `ImmutableDictionary`-with-comparer around using the `MSBuildNameIgnoreCaseComparer`, which is the primary comparer used for all these dictionaries in MSBuild. `OrdinalIgnoreCase` is also used and gets a special case too. To make the type system happier, it was easiest to drop the `K` generic parameter to the type and hard-code it to `string`--all internal non-test uses were already doing that.
Put under a change wave
Put under a change wave
Put under a change wave
This prevents a single .sln and a single .slnf from causing an ambiguous .sln error. It prioritizes building the .slnf when we would build a .sln but it is absent.
Simplify search for which files might be solution files, project files, or solution filter files. Also should be *slightly* more performant, but that isn't a serious consideration here.
Enable change waves. Added static ChangeWaves class that contains AreFeaturesEnabled and ApplyChangeWave. Documentation is in the wiki.
This reverts commit a5a2d70.
benvillalobos
approved these changes
Sep 25, 2020
Member
benvillalobos
left a comment
There was a problem hiding this comment.
LGTM.
Includes a fix to one of the issues referenced here: #5756
10 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
I've been trying to come up with a good way to test the full drive globs PR, but the way I can think of to test opting out of that is to make sure we scan the whole drive, which would take rather a long time, so it wouldn't be good for a test. Alternative?
The core count has a different problem: the behavior should be identical if you have <= 32 cores whether or not the change wave is used, and it would be very confusing to have the test sometimes work and sometimes fail.
One idea I came up with is to log a message when using the new core detection logic. For the full drive scan, we could theoretically put in something that logs when we are doing a full drive scan in a test and suppresses it. It would be nice to have a TestSystem that returns very little when you try to look through its drive, but unless that already exists, I'd say the cost outweighs the benefits here.