Conversation
|
(rust-highfive has picked a reviewer for you, use r? to override) |
This comment has been minimized.
This comment has been minimized.
|
Seems pretty surprising for the compiler to crash on that test case. I can't reproduce that here, obviously. I guess I can drop that one from the patch though. |
Bug rust-lang#47163 points out that some debug tests were disabled. This patch re-enables many of them for gdb (I don't know enough about recent versions of lldb to make this change there). One test that only works for a relatively new gdb, due to the use of DW_ATE_UTF, is moved to a version-specific test. In a couple of spot, I simplified the check expression where I thought that it made sense to do so -- sometimes it isn't very worthwhile to match exactly what is emitted, particularly where this has changed over time. I tested this with both a DW_ATE_UTF-capable gdb, and an older version.
The compiler crashed in CI when compiling drop-locations.rs, so restore the ignore-test marker there.
edd74cf to
553e106
Compare
|
Rebased + reverted the one failing bit. |
|
@bors r+ rollup=iffy |
|
📌 Commit 553e106 has been approved by |
|
⌛ Testing commit 553e106 with merge 38a30ac11295d9d59aee3e560fb2abc7bbc3342a... |
|
💔 Test failed - checks-actions |
|
The job Click to see the possible cause of the failure (guessed by this bot) |
|
⌛ Testing commit 553e106 with merge 5891376e9c37035bac0606cca4af9c39aa213b30... |
|
The job Click to see the possible cause of the failure (guessed by this bot) |
|
💔 Test failed - checks-actions |
|
I did a bors resync and this seems to have gotten reapproved, sorry for the noise @bors r- |
|
ping from triage: FYI: when a PR is ready for review, send a message containing |
|
☔ The latest upstream changes (presumably #95685) made this pull request unmergeable. Please resolve the merge conflicts. |
|
Closing this due to inactivity |
Bug #47163 points out that some debug tests were disabled. This patch
re-enables many of them for gdb (I don't know enough about recent
versions of lldb to make this change there). One test that only works
for a relatively new gdb, due to the use of DW_ATE_UTF, is moved to a
version-specific test. In a couple of spot, I simplified the check
expression where I thought that it made sense to do so -- sometimes it
isn't very worthwhile to match exactly what is emitted, particularly
where this has changed over time. I tested this with both a
DW_ATE_UTF-capable gdb, and an older version.