Fix make check-stage2 by fixing deps of exported syntax cfail tests.
#14014
Conversation
This file contains 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
Our `make check-stageN` for all N (including N=2) was not working, because there is an unspecified dependence from the cfail tests on certain crates that export syntax. We need to encode that dependence in some manner. The quick hack is to add `$$(CSREQ$(1)_T_$(3)_H_$(3)) $$(SREQ$(1)_T_$(2)_H_$(3))` to the dependency list for *all* of the cfail tests. But I would prefer to not force the make logic to build all of the external crates before it gets to exercise the cfail tests. So, here is how this PR fixes the problem more properly: * Add `CTEST_DEPS_cfail-full` that, analogous to `CTEST_DEPS_rpass-full`, has a dependency on `$$(CSREQ$(1)_T_$(3)_H_$(3)) $$(SREQ$(1)_T_$(2)_H_$(3))`. `CTEST_DEPS_cfail` is left unchanged (and thus continues to match the structure of `CTEST_DEPS_rpass`). * Added all of the other ingredients for `cfail-full` family of make logic. * moved cfail tests that rely on macros in separate crates to separate compile-fail-fulldeps/ directory. ---- (Landing this may or not be a prerequisite for landing PR rust-lang#14000. In any case, this is a pretty isolated change and a net improvement in the overall structure for the tests, IMO, at least given what we currently have in the makefiles.)
|
r? anyone interested/semi-knowledgeable in our make architecture. |
|
This looks very eerily similar to #13832 ... |
|
@alexcrichton d'oh. Sorry. Let me see where bors is on this race. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Our
make check-stageNfor all N (including N=2) was not working,because there is an unspecified dependence from the cfail tests on
certain crates that export syntax.
We need to encode that dependence in some manner. The quick hack is
to add
$$(CSREQ$(1)_T_$(3)_H_$(3)) $$(SREQ$(1)_T_$(2)_H_$(3))to thedependency list for all of the cfail tests. But I would prefer to
not force the make logic to build all of the external crates before it
gets to exercise the cfail tests.
So, here is how this PR fixes the problem more properly:
CTEST_DEPS_cfail-fullthat, analogous toCTEST_DEPS_rpass-full, has a dependency on$$(CSREQ$(1)_T_$(3)_H_$(3)) $$(SREQ$(1)_T_$(2)_H_$(3)).CTEST_DEPS_cfailis left unchanged (and thus continues tomatch the structure of
CTEST_DEPS_rpass).cfail-fullfamily of makelogic.
separate compile-fail-fulldeps/ directory.
(Landing this may or not be a prerequisite for landing PR #14000. In
any case, this is a pretty isolated change and a net improvement in
the overall structure for the tests, IMO, at least given what we
currently have in the makefiles.)