Remove size_of == 1 case from fill specialization. - #87892
Conversation
|
r? @yaahc (rust-highfive has picked a reviewer for you, use r? to override) |
There was a problem hiding this comment.
Should be fine as an immediate fix to remove the UB, but I think it's still better to explicitly tell the compiler that it's dealing with a memset when we can rather than relying on optimizers. Currently it helps at least one other backend. And even llvm will have to do less work if we just tell it.
|
@bors r=the8472 rollup=never |
|
📌 Commit 3838301 has been approved by |
|
⌛ Testing commit 3838301 with merge 9910329347e84f297a0281952e6857c145e043a2... |
|
💥 Test timed out |
|
@bors retry |
|
@m-ou-se Miri actually runs libcore tests (https://github.com/rust-lang/miri-test-libstd), so it'd be great if you could add your test case from #87891 to the test suite -- doesn't have to be in this PR though. |
|
I think github had some hiccups today. @bors retry |
|
⌛ Testing commit 3838301 with merge 0443466ce9adb5af19c2273798e6dcbb36a7dbf4... |
|
The job Click to see the possible cause of the failure (guessed by this bot) |
|
💔 Test failed - checks-actions |
|
@bors retry |
|
☀️ Test successful - checks-actions |
Fixes #87891
See discussion on Zulip.