fix remaining problem with multi-statement inlining#14230
fix remaining problem with multi-statement inlining#14230dlang-bot merged 2 commits intodlang:masterfrom
Conversation
|
Thanks for your pull request, @WalterBright! Bugzilla referencesYour PR doesn't reference any Bugzilla issue. If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog. Testing this PR locallyIf you don't have a local development environment setup, you can use Digger to test this PR: dub run digger -- build "master + dmd#14230" |
f44dfa5 to
020ed77
Compare
|
@rainers you are probably the most knowledgeable person on exceptions for mscoff32. Can you please take a look at this and point me at something I can fix? The failure is most likely due to something being inlined in the runtime library, as testline.d itself is not inlining anything according to my tests. |
It uses the same exception mechanism as the OMF version, i.e. deh_win32. As far as I can tell from stepping through a build with the debugger (debug info is pretty broken in optimized builds), the problem is in |
|
Thanks, @rainers ! I will check this out. |
|
Which |
See dlang/druntime#3853 (comment) BTW: AFAICT the win*.mak files were left in their dire state mostly to not interrupt your workflow, especially with respect to using VC10. |
|
Finally, I can repro the problem! Yay! |
3331592 to
41f3051
Compare
|
buildkite seems to hang. |
|
The 3 buildkite errors are all of the form: But I have no idea why buildkite fails to build the library while every other test that builds the library succeeds. I do not know what is different about the buildkite builds. |
15a0d00 to
1b8f146
Compare
1b8f146 to
fe8ab7e
Compare
fe8ab7e to
f55393a
Compare
f55393a to
8dfa638
Compare
failing runnable/testline.d