Downgrade bundled jemalloc version#30985
Conversation
|
(rust_highfive has picked a reviewer for you, use r? to override) |
|
@bors r+ Might as well run the bugs that we didn't run into. |
|
📌 Commit ab758e5 has been approved by |
|
@bors p=1 |
|
⌛ Testing commit ab758e5 with merge 6834008... |
|
💔 Test failed - auto-win-gnu-64-nopt-t |
|
Hm, this is... interesting. We seem to be in a bit of a quandry with respect to jemalloc on Windows. I suspect that this is jemalloc/jemalloc#310 coming to haunt us again. In the past the All in all, it appears that Thoughts @rust-lang/tools? Opinions on disabling jemalloc for Windows temporarily until we can re-upgrade again? |
|
Well, if it breaks stuff, we should disable it. Are we gonna lose on some important benchmark because of this? |
|
I suspect we'll looks on some benchmark as jemalloc is probably faster than the system allocator for at least some workloads, but I also agree that it's likely more important to work at all! |
|
Seems to me we should disable jemalloc on windows, yes. It'd be great if we can make a corresponding Rust issue with the details and citations you've dumped on this thread, so that we have a central place to track this. (Would this mean that we would enable the newer jemalloc elsewhere? I'm guess not, because jemalloc/jemalloc#315 doesn't appear to be specific to Windows.) |
ab758e5 to
840a1d7
Compare
|
Ok, I've filed an issue and disabled jemalloc on windows-gnu. r? @brson |
|
@bors r+ p=1 |
|
📌 Commit 840a1d7 has been approved by |
|
⌛ Testing commit 840a1d7 with merge 05756dd... |
|
💔 Test failed - auto-win-gnu-32-nopt-t |
840a1d7 to
66cdd85
Compare
|
⌛ Testing commit 66cdd85 with merge 5a50f7f... |
|
💔 Test failed - auto-win-gnu-64-opt |
We've been seeing a lot of timeouts in tests on the bots and investigation ended pointing to jemalloc/jemalloc#315 as the culprit. Unfortunately it looks like that doesn't seem to have a fix on the way soon, so let's temporarily downgrade back to the previous version of jemalloc we were using (where rust-lang#30434 was the most recent upgrade)
66cdd85 to
884de56
Compare
We've been seeing a lot of timeouts in tests on the bots and investigation ended pointing to jemalloc/jemalloc#315 as the culprit. Unfortunately it looks like that doesn't seem to have a fix on the way soon, so let's temporarily downgrade back to the previous version of jemalloc we were using (where #30434 was the most recent upgrade)
We've been seeing a lot of timeouts in tests on the bots and investigation ended
pointing to jemalloc/jemalloc#315 as the culprit. Unfortunately it looks like
that doesn't seem to have a fix on the way soon, so let's temporarily downgrade
back to the previous version of jemalloc we were using (where #30434 was the
most recent upgrade)