Skip documentation for tier 2 targets on dist-x86_64-apple-darwin #90100
Conversation
rust-highfive
added
the
S-waiting-on-review
|
Let me actually get this into a try build to test it |
|
@bors try |
|
⌛ Trying commit 3eae37c4c6795401a15a07028d607a690842ccad with merge a2b2936730df129423af44dab918d380df6ccdd4... |
This comment has been minimized.
This comment has been minimized.
|
💔 Test failed - checks-actions |
bors
added
S-waiting-on-author
Mark-Simulacrum
force-pushed
the
speed-macos-ci
branch
from
3eae37c to
5503dd9
Compare
|
@bors try |
|
⌛ Trying commit 5503dd9 with merge c18b3fe6f6448517cfb25f8facf58bfdb46016ad... |
|
☀️ Try build successful - checks-actions |
|
@bors try |
|
⌛ Trying commit 9296efe with merge 8a9d748d49be179776d8e74ef8b372c75449e57e... |
|
☀️ Try build successful - checks-actions |
|
r=me once the try commit is removed. |
|
Well, try commits so far are yielding regressions in performance I think so deeper investigation necessary :) |
|
Well the r=me was mostly about the code change itself 😅 |
|
Looks like this currently adds ~20 minutes to the time across try runs which seem to be just noisy neighbors or something -- all the times for other jobs went up. However, I've checked that the artifacts produced look good and we are cutting out the fairly costly docs building. So I think we should probably go ahead and merge this; unfortunately we don't currently really track CI times anywhere, so it'll be a little hard to get a sense of whether this is an improvement, but I think the potential time savings are worth it. I also have an idea for how to track those that I'm going to experiment with a little today, but it'll be able to back-collect trivially. |
Mark-Simulacrum
added
S-waiting-on-review
|
OK, I have a local setup for monitoring CI times now that should hopefully work well enough that we can detect changes caused by this PR (given the expected multi-minute impact). I'm going to go ahead and approve (per earlier r=me) and will check in on the auto branch builds in a few days to see what the effect of this PR is. My guess is it'll help, but we'll see. @bors r=pietroalbini For the record, currently seeing (rolling 32 commit median time):
|
|
📌 Commit 9296efe has been approved by |
bors
added
S-waiting-on-bors
…ietroalbini Skip documentation for tier 2 targets on dist-x86_64-apple-darwin I don't have an easy way to test this locally, but I believe it should work. Based on one log result should shave ~14 minutes off the dist-x86_64-apple builder (doesn't help with aarch64 dist or x86_64 test builder, so not actually decreasing total CI time most likely). r? `@pietroalbini`
|
Got an "attempt to subtract with overfow" error during std documentation in a rollup, could this be related to this pr? Edit: nevermind, #90042 (comment) had the same problem and it didn't contain this pr |
…ietroalbini Skip documentation for tier 2 targets on dist-x86_64-apple-darwin I don't have an easy way to test this locally, but I believe it should work. Based on one log result should shave ~14 minutes off the dist-x86_64-apple builder (doesn't help with aarch64 dist or x86_64 test builder, so not actually decreasing total CI time most likely). r? `@pietroalbini`
…ietroalbini Skip documentation for tier 2 targets on dist-x86_64-apple-darwin I don't have an easy way to test this locally, but I believe it should work. Based on one log result should shave ~14 minutes off the dist-x86_64-apple builder (doesn't help with aarch64 dist or x86_64 test builder, so not actually decreasing total CI time most likely). r? ``@pietroalbini``
…askrgr Rollup of 6 pull requests Successful merges: - rust-lang#89558 (Add rustc lint, warning when iterating over hashmaps) - rust-lang#90100 (Skip documentation for tier 2 targets on dist-x86_64-apple-darwin) - rust-lang#90155 (Fix alignment of method headings for scannability) - rust-lang#90162 (Mark `{array, slice}::{from_ref, from_mut}` as const fn) - rust-lang#90221 (Fix ICE when forgetting to `Box` a parameter to a `Self::func` call) - rust-lang#90234 (Temporarily turn overflow checks off for rustc-rayon-core) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
…bini Move back to linux builder on try builds Apparently deleted the wrong line when trying to revert changes to try in rust-lang#90100 which I now see still contains the do not merge commit -- maybe I forgot to force push the local changes I had pending or something. r? `@pietroalbini`
I don't have an easy way to test this locally, but I believe it should work. Based on one log result should shave ~14 minutes off the dist-x86_64-apple builder (doesn't help with aarch64 dist or x86_64 test builder, so not actually decreasing total CI time most likely).
r? @pietroalbini