Encode only MIR reachable from other crates #115306

tmiasko · 2023-08-28T07:59:09Z

Only reachable items might participate in the code generation in the
downstream crates. Omit redundant optimized MIR of unreachable items
from a crate metadata.

Additionally, include reachable closures in reachable set, so that
unreachable closures can be omitted on the same basis.

tmiasko · 2023-08-28T08:02:11Z

@bors try @rust-timer queue

bors · 2023-08-28T08:02:21Z

⌛ Trying commit 7a2b2ca0a9f3c45e766138547f81d4a130ba09d4 with merge 524a72e9fbc25be53b017952c3f2520a769c65c0...

tmiasko · 2023-08-28T09:44:32Z

@bors try @rust-timer queue

bors · 2023-08-28T09:44:42Z

⌛ Trying commit 18f8aa6acb7bdb7bbee9ddd47b74d6453c879a08 with merge 8befefe0eeb75ffb3138d26537e90703d068ea2a...

bors · 2023-08-28T10:52:57Z

💔 Test failed - checks-actions

tmiasko · 2023-08-28T11:09:21Z

@rust-timer build 524a72e9fbc25be53b017952c3f2520a769c65c0

tmiasko · 2023-08-28T18:41:11Z

Toolchain	Bytes
`41cb42a`	490193897
524a72e9fbc25be53b017952c3f2520a769c65c0	479937961

rust-timer · 2023-08-28T23:55:04Z

Finished benchmarking commit (524a72e9fbc25be53b017952c3f2520a769c65c0): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.4%, 2.6%]	30
Regressions ❌ (secondary)	1.5%	[0.4%, 9.2%]	10
Improvements ✅ (primary)	-1.2%	[-2.6%, -0.3%]	11
Improvements ✅ (secondary)	-0.9%	[-1.1%, -0.8%]	3
All ❌✅ (primary)	0.2%	[-2.6%, 2.6%]	41

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	1.4%	[0.9%, 1.9%]	5
Regressions ❌ (secondary)	2.1%	[0.7%, 4.6%]	7
Improvements ✅ (primary)	-1.1%	[-1.5%, -0.6%]	5
Improvements ✅ (secondary)	-0.6%	[-0.6%, -0.6%]	2
All ❌✅ (primary)	0.2%	[-1.5%, 1.9%]	10

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	1.8%	[1.4%, 3.0%]	7
Regressions ❌ (secondary)	5.0%	[3.2%, 9.6%]	4
Improvements ✅ (primary)	-1.7%	[-2.1%, -1.3%]	4
Improvements ✅ (secondary)	-2.4%	[-2.5%, -2.4%]	2
All ❌✅ (primary)	0.6%	[-2.1%, 3.0%]	11

Binary size

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-1.1%	[-7.1%, -0.0%]	87
Improvements ✅ (secondary)	-1.0%	[-3.7%, -0.0%]	37
All ❌✅ (primary)	-1.1%	[-7.1%, -0.0%]	87

Bootstrap: 631.418s -> 630.932s (-0.08%)
Artifact size: 316.30 MiB -> 315.75 MiB (-0.17%)

tmiasko · 2023-08-30T06:36:53Z

@bors try @rust-timer queue

bors · 2023-08-30T06:37:02Z

⌛ Trying commit 945cde202a1018cf8f93fea755db3da165f355bc with merge 69b9a1d784d7997066edd67bfea933c18927502e...

bors · 2023-08-30T07:47:19Z

☀️ Try build successful - checks-actions
Build commit: 69b9a1d784d7997066edd67bfea933c18927502e (69b9a1d784d7997066edd67bfea933c18927502e)

cjgillot · 2023-09-07T20:00:20Z

compiler/rustc_metadata/src/rmeta/encoder.rs

@@ -1580,9 +1585,10 @@ impl<'a, 'tcx> EncodeContext<'a, 'tcx> {
        }

        let tcx = self.tcx;
+        let reachable_set = tcx.reachable_set(());

        let keys_and_jobs = tcx.mir_keys(()).iter().filter_map(|&def_id| {


Should we filter more aggressively using a .filter(|def_id| reachable_set.contains(def_id))?

reachable_set is used to filter the optimized MIR only. It might still be necessary to encode MIR for CTFE.

That's my question: should we filter MIR for CTFE too according to the same rule?

The reachable set cannot be used for the purpose. It contains items externally reachable in the linkage sense, or that could be code generated while building other crates, but it doesn't have much to say about items which might evaluated in other crates. Implementing this would require a separate analysis (I assume we don't want to include those in reachable set since that would be pessimization).

Could you add a line explaining the logic around reachable_set and this exception to the doc-comment on should_encode_mir?
r=me then

Only reachable items might participate in the code generation in the downstream crates. Omit redundant optimized MIR of unreachable items from a crate metadata. Additionally, include reachable closures in reachable set, so that unreachable closures can be omitted on the same basis.

tmiasko · 2023-09-10T12:55:03Z

@bors r=cjgillot

bors · 2023-09-10T12:55:05Z

📌 Commit d99333e has been approved by cjgillot

It is now in the queue for this repository.

bors · 2023-09-10T12:55:14Z

⌛ Testing commit d99333e with merge 65b2ffd...

…llot Encode only MIR reachable from other crates Only reachable items might participate in the code generation in the downstream crates. Omit redundant optimized MIR of unreachable items from a crate metadata. Additionally, include reachable closures in reachable set, so that unreachable closures can be omitted on the same basis.

bors · 2023-09-10T14:42:49Z

☀️ Test successful - checks-actions
Approved by: cjgillot
Pushing 65b2ffd to master...

bors · 2023-09-10T14:43:49Z

👀 Test was successful, but fast-forwarding failed: 422 Update is not a fast forward

tmiasko · 2023-09-10T14:54:08Z

@bors retry

bors · 2023-09-10T16:31:14Z

⌛ Testing commit d99333e with merge 7418413...

rust-timer · 2023-09-10T17:05:36Z

Finished benchmarking commit (65b2ffd): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

Next Steps: If you can justify the regressions found in this perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please open an issue or create a new PR that fixes the regressions, add a comment linking to the newly created issue or PR, and then add the perf-regression-triaged label to this PR.

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.8%	[0.2%, 2.5%]	23
Regressions ❌ (secondary)	1.4%	[0.4%, 9.2%]	12
Improvements ✅ (primary)	-2.0%	[-2.8%, -0.8%]	6
Improvements ✅ (secondary)	-1.0%	[-1.0%, -0.9%]	3
All ❌✅ (primary)	0.2%	[-2.8%, 2.5%]	29

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	1.5%	[0.7%, 2.4%]	8
Regressions ❌ (secondary)	3.5%	[1.3%, 6.3%]	5
Improvements ✅ (primary)	-1.5%	[-1.6%, -1.3%]	2
Improvements ✅ (secondary)	-2.1%	[-3.9%, -0.5%]	4
All ❌✅ (primary)	0.9%	[-1.6%, 2.4%]	10

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	2.0%	[1.1%, 3.4%]	7
Regressions ❌ (secondary)	4.7%	[2.1%, 9.1%]	5
Improvements ✅ (primary)	-1.7%	[-1.8%, -1.6%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.9%	[-1.8%, 3.4%]	10

Binary size

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-1.2%	[-7.5%, -0.0%]	78
Improvements ✅ (secondary)	-1.2%	[-4.9%, -0.0%]	37
All ❌✅ (primary)	-1.2%	[-7.5%, -0.0%]	78

Bootstrap: 629.364s -> 629.035s (-0.05%)
Artifact size: 317.94 MiB -> 317.47 MiB (-0.15%)

bors · 2023-09-10T18:18:48Z

☀️ Test successful - checks-actions
Approved by: cjgillot
Pushing 7418413 to master...

rust-timer · 2023-09-10T19:47:11Z

Finished benchmarking commit (7418413): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

Next Steps: If you can justify the regressions found in this perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please open an issue or create a new PR that fixes the regressions, add a comment linking to the newly created issue or PR, and then add the perf-regression-triaged label to this PR.

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.8%	[0.3%, 2.4%]	15
Regressions ❌ (secondary)	1.9%	[0.3%, 9.1%]	7
Improvements ✅ (primary)	-1.3%	[-2.7%, -0.4%]	12
Improvements ✅ (secondary)	-0.9%	[-1.2%, -0.7%]	5
All ❌✅ (primary)	-0.1%	[-2.7%, 2.4%]	27

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	1.7%	[0.9%, 2.4%]	7
Regressions ❌ (secondary)	4.5%	[2.4%, 6.1%]	3
Improvements ✅ (primary)	-1.4%	[-2.4%, -0.6%]	4
Improvements ✅ (secondary)	-1.3%	[-2.9%, -0.6%]	4
All ❌✅ (primary)	0.6%	[-2.4%, 2.4%]	11

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	2.0%	[1.3%, 3.4%]	6
Regressions ❌ (secondary)	3.3%	[1.7%, 8.7%]	7
Improvements ✅ (primary)	-2.3%	[-2.3%, -2.3%]	1
Improvements ✅ (secondary)	-2.4%	[-2.4%, -2.4%]	1
All ❌✅ (primary)	1.4%	[-2.3%, 3.4%]	7

Binary size

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-1.2%	[-7.5%, -0.0%]	79
Improvements ✅ (secondary)	-1.2%	[-4.9%, -0.0%]	37
All ❌✅ (primary)	-1.2%	[-7.5%, -0.0%]	79

Bootstrap: 631.343s -> 629.704s (-0.26%)
Artifact size: 318.07 MiB -> 317.61 MiB (-0.15%)

pnkfelix · 2023-09-13T20:17:27Z

the big (>1%) primary regressions were to three check-incr-unchanged cases: cranelift-codegen-0.82.1, html5ever-0.26.0, and hyper-0.14.18
the regressions seem unfortunate, but tolerable given the improvement to linked artifact sizes
marking as triaged

@rustbot label: +perf-regression-triaged

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Aug 28, 2023

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Aug 28, 2023

tmiasko force-pushed the encode-reachable-mir branch from 7a2b2ca to 18f8aa6 Compare August 28, 2023 09:12

This comment has been minimized.

Sign in to view

bors added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Aug 28, 2023

This comment has been minimized.

Sign in to view

rustbot added perf-regression Performance regression. and removed S-waiting-on-perf Status: Waiting on a perf run to be completed. labels Aug 28, 2023

tmiasko force-pushed the encode-reachable-mir branch from 18f8aa6 to 945cde2 Compare August 30, 2023 06:33

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Aug 30, 2023

This comment has been minimized.

Sign in to view

tmiasko added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Sep 7, 2023

cjgillot reviewed Sep 7, 2023

View reviewed changes

cjgillot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Sep 9, 2023

tmiasko force-pushed the encode-reachable-mir branch from b5b696c to 77bde92 Compare September 10, 2023 11:16

This comment has been minimized.

Sign in to view

tmiasko force-pushed the encode-reachable-mir branch from 77bde92 to d99333e Compare September 10, 2023 11:55

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Sep 10, 2023

bors added the merged-by-bors This PR was explicitly merged by bors. label Sep 10, 2023

tmiasko removed the merged-by-bors This PR was explicitly merged by bors. label Sep 10, 2023

bors added the merged-by-bors This PR was explicitly merged by bors. label Sep 10, 2023

bors merged commit 7418413 into rust-lang:master Sep 10, 2023
12 checks passed

rustbot added this to the 1.74.0 milestone Sep 10, 2023

tmiasko deleted the encode-reachable-mir branch September 10, 2023 18:19

rustbot added the perf-regression-triaged The performance regression has been triaged. label Sep 13, 2023

saethlin mentioned this pull request Mar 23, 2024

improve codegen of fmt_num to delete unreachable panic #122770

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Encode only MIR reachable from other crates #115306

Encode only MIR reachable from other crates #115306

tmiasko commented Aug 28, 2023 •

edited

tmiasko commented Aug 28, 2023

This comment has been minimized.

bors commented Aug 28, 2023

tmiasko commented Aug 28, 2023

This comment has been minimized.

bors commented Aug 28, 2023

bors commented Aug 28, 2023

This comment has been minimized.

tmiasko commented Aug 28, 2023

This comment has been minimized.

tmiasko commented Aug 28, 2023

rust-timer commented Aug 28, 2023

tmiasko commented Aug 30, 2023

This comment has been minimized.

bors commented Aug 30, 2023

This comment has been minimized.

bors commented Aug 30, 2023

cjgillot Sep 7, 2023

tmiasko Sep 8, 2023

cjgillot Sep 9, 2023

tmiasko Sep 9, 2023

cjgillot Sep 9, 2023

This comment has been minimized.

tmiasko commented Sep 10, 2023

bors commented Sep 10, 2023

bors commented Sep 10, 2023

bors commented Sep 10, 2023

bors commented Sep 10, 2023

tmiasko commented Sep 10, 2023

bors commented Sep 10, 2023

rust-timer commented Sep 10, 2023

bors commented Sep 10, 2023

rust-timer commented Sep 10, 2023

pnkfelix commented Sep 13, 2023

Encode only MIR reachable from other crates #115306

Encode only MIR reachable from other crates #115306

Conversation

tmiasko commented Aug 28, 2023 • edited

tmiasko commented Aug 28, 2023

This comment has been minimized.

bors commented Aug 28, 2023

tmiasko commented Aug 28, 2023

This comment has been minimized.

bors commented Aug 28, 2023

bors commented Aug 28, 2023

This comment has been minimized.

tmiasko commented Aug 28, 2023

This comment has been minimized.

tmiasko commented Aug 28, 2023

rust-timer commented Aug 28, 2023

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

tmiasko commented Aug 30, 2023

This comment has been minimized.

bors commented Aug 30, 2023

This comment has been minimized.

bors commented Aug 30, 2023

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

This comment has been minimized.

tmiasko commented Sep 10, 2023

bors commented Sep 10, 2023

bors commented Sep 10, 2023

bors commented Sep 10, 2023

bors commented Sep 10, 2023

tmiasko commented Sep 10, 2023

bors commented Sep 10, 2023

rust-timer commented Sep 10, 2023

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

bors commented Sep 10, 2023

rust-timer commented Sep 10, 2023

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

pnkfelix commented Sep 13, 2023

tmiasko commented Aug 28, 2023 •

edited