[WIP] Re-enable the early otherwise branch optimization #121397

DianQK · 2024-02-21T14:56:52Z

Fixes #95162. Fixes #119014. Fixes #117970.

An invalid enum discriminant can come from anywhere. We have to check to see if all successors contain the discriminant statement.

It should not be UB that we pass in an invalid enum discriminant when calling a function, this is more like LLVM's poison value. UB only when used. Although miri would consider the following code to be UB. (It's fine for miri.)

https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=18602870aaeb07cbdf7dfcd2c28961a2

I extended the scenario with scalars and the same target values.

r? compiler

rustbot · 2024-02-21T14:57:02Z

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

michaelwoerister · 2024-02-21T15:00:47Z

r? mir-opt

bors · 2024-02-22T19:24:23Z

☔ The latest upstream changes (presumably #121370) made this pull request unmergeable. Please resolve the merge conflicts.

DianQK · 2024-02-29T05:05:01Z

@bors try @rust-timer queue

bors · 2024-02-29T11:14:35Z

⌛ Trying commit d5c6f73 with merge 7a8b62c...

… r=<try> Re-enable the early otherwise branch optimization Fixes rust-lang#95162. Fixes rust-lang#119014. Fixes rust-lang#117970. An invalid enum discriminant can come from anywhere. We have to check to see if all successors contain the discriminant statement. It should not be UB that we pass in an invalid enum discriminant when calling a function, this is more like LLVM's poison value. UB only when used. Although miri would consider the following code to be UB. (It's fine for miri.) https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=18602870aaeb07cbdf7dfcd2c28961a2 I extended the scenario with scalars and the same target values. r? compiler

bors · 2024-02-29T12:51:31Z

☀️ Try build successful - checks-actions
Build commit: 7a8b62c (7a8b62c30d4f0ad6b88fa2bffff86fa70e668cd4)

rust-timer · 2024-02-29T14:26:39Z

Finished benchmarking commit (7a8b62c): 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)	1.5%	[0.3%, 2.6%]	2
Regressions ❌ (secondary)	1.2%	[0.9%, 2.6%]	7
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.7%	[-2.0%, -1.4%]	2
All ❌✅ (primary)	1.5%	[0.3%, 2.6%]	2

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)	3.8%	[3.8%, 3.8%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.0%	[-0.0%, -0.0%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.9%	[-0.0%, 3.8%]	2

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)	2.5%	[2.5%, 2.5%]	1
Regressions ❌ (secondary)	3.2%	[3.2%, 3.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.5%	[2.5%, 2.5%]	1

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)	0.3%	[0.3%, 0.3%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.3%	[0.3%, 0.3%]	1

Bootstrap: 650.923s -> 650.175s (-0.11%)
Artifact size: 311.19 MiB -> 311.09 MiB (-0.03%)

… target values are the same

bors · 2024-03-03T21:51:30Z

⌛ Trying commit 73ef0c0 with merge a4a739a...

… r=<try> Re-enable the early otherwise branch optimization Fixes rust-lang#95162. Fixes rust-lang#119014. Fixes rust-lang#117970. An invalid enum discriminant can come from anywhere. We have to check to see if all successors contain the discriminant statement. It should not be UB that we pass in an invalid enum discriminant when calling a function, this is more like LLVM's poison value. UB only when used. Although miri would consider the following code to be UB. (It's fine for miri.) https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=18602870aaeb07cbdf7dfcd2c28961a2 I extended the scenario with scalars and the same target values. r? compiler

bors · 2024-03-03T23:19:00Z

☀️ Try build successful - checks-actions
Build commit: a4a739a (a4a739ad3555a822cf38daebb07f963e3cf8a3c2)

rust-timer · 2024-03-04T00:42:14Z

Finished benchmarking commit (a4a739a): comparison URL.

Overall result: ❌ regressions - no 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.

@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)	2.7%	[2.7%, 2.7%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.7%	[2.7%, 2.7%]	1

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)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.1%	[-2.6%, -1.6%]	2
Improvements ✅ (secondary)	-2.7%	[-4.4%, -2.0%]	7
All ❌✅ (primary)	-2.1%	[-2.6%, -1.6%]	2

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)	2.7%	[2.7%, 2.7%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.7%	[2.7%, 2.7%]	1

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)	-0.1%	[-0.1%, -0.1%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.1%	[-0.1%, -0.1%]	1

Bootstrap: 649.358s -> 650.059s (0.11%)
Artifact size: 175.04 MiB -> 175.07 MiB (0.02%)

DianQK · 2024-03-04T01:13:08Z

r? cjgillot

rustbot · 2024-03-04T01:13:10Z

Could not assign reviewer from: cjgillot.
User(s) cjgillot are either the PR author, already assigned, or on vacation, and there are no other candidates.
Use r? to specify someone else to assign.

cjgillot

I have made a first pass over the implementation. This pass is very hard to review, so I may need to come back several times to it.

cjgillot · 2024-03-10T13:25:47Z

compiler/rustc_mir_transform/src/early_otherwise_branch.rs

+            // invalid value, which is UB.
+            // In order to fix this, we would either need to show that the discriminant computation of
+            // `place` is computed in all branches.
+            // So we need the `otherwise` branch has no statements and an unreachable terminator.


The opt would be correct if we compute the discriminant in the otherwise branch like all other ones, wouldn't it?

Yes. But I think it could make this pass become complicated. So I decided to keep it simple first. This way, at least the pass can start to do work, and iterate over it in follow-on updates.

cjgillot · 2024-03-10T13:26:54Z

compiler/rustc_mir_transform/src/early_otherwise_branch.rs

 #[derive(Debug)]
 struct OptimizationData<'tcx> {
    destination: BasicBlock,
    child_place: Place<'tcx>,
    child_ty: Ty<'tcx>,
    child_source: SourceInfo,
+    hoist_discriminant: bool,
+    same_target_value: Option<u128>,


Could you add docs for those fields?

cjgillot · 2024-03-10T13:27:59Z

compiler/rustc_mir_transform/src/early_otherwise_branch.rs

+        if child_targets.otherwise() != destination {
+            return None;
+        }
+        // Make sure there are only two branches.


Should this comment be moved?

cjgillot · 2024-03-10T13:29:03Z

compiler/rustc_mir_transform/src/early_otherwise_branch.rs

 #[derive(Debug)]
 struct OptimizationData<'tcx> {
    destination: BasicBlock,
    child_place: Place<'tcx>,
    child_ty: Ty<'tcx>,
    child_source: SourceInfo,
+    hoist_discriminant: bool,
+    same_target_value: Option<u128>,
 }

 fn evaluate_candidate<'tcx>(


I have a lot of trouble understanding how this function is structured and what it evaluates exactly.

cjgillot · 2024-03-10T13:30:11Z

compiler/rustc_mir_transform/src/early_otherwise_branch.rs

-    if destination != targets.otherwise() {
-        return false;
+    if hoist_discriminant {
+        // ...assign the discriminant of `place` in that statement


Nit: the ... pattern is lost since you moved code into evaluate_candidate.

erikdesjardins · 2024-03-10T21:10:17Z

It should not be UB that we pass in an invalid enum discriminant when calling a function, this is more like LLVM's poison value. UB only when used.

I don't think it's an issue for this PR's implementation (since it's conservative), but passing an invalid value to a function is UB. This is very explicit in the reference:

Producing an invalid value, even in private fields and locals. "Producing" a value happens any time a value is assigned to or read from a place, passed to a function/primitive operation or returned from a function/primitive operation. The following values are invalid (at their respective type):

[...]

A discriminant in an enum not included in the type definition.

The unsoundness described in #95162 occurs for exactly this reason--the old implementation would hoist the discriminant(Q) operation, causing it to execute on code paths where it wouldn't before, introducing new instances of this kind of UB. If the discriminant(Q) operation merely returned something poison-like, instead of invoking UB when you pass an invalid value to it, that wouldn't be an issue.

DianQK · 2024-03-11T01:20:50Z

I have made a first pass over the implementation. This pass is very hard to review, so I may need to come back several times to it.

Sorry for that, it's probably because I have new findings and update to this PR. I can split this up into multiple PRs. The current commit history should reflect that as well.

Major changes (PRs that can be split):

Fix the unsound issue and re-enable this pass.
Add scalar support.
Add pattern with the same target value.

DianQK · 2024-03-11T01:26:07Z

Yes. I plan to submit a series of PRs with these reviews.

DianQK · 2024-03-11T01:29:45Z

@erikdesjardins Thanks for the explanation. I will document this in a new PR.
I want to leave a FIXME and re-enable it. Rather than disable it and making it difficult to iterate.

DianQK · 2024-03-12T03:07:54Z

Very interesting, I broke this PR with #120268.
Since we haven't enabled this pass yet, I don't think this is a regression.

See https://play.rust-lang.org/?version=nightly&mode=release&edition=2021&gist=7606620b8e2ab84fe0c6e06b4ec3e7c4.
We change switchInt(move _3) -> [0: bb5, otherwise: bb1] to switchInt(move _3) -> [0: bb5, 1: bb1, otherwise: bb7];. This is not an if else statement now.

Sometimes we fail to make better with better facts. :(
But I can fix it and let me refine this pass step by step. :3

DianQK · 2024-03-12T14:44:08Z

#122387 is the first PR.

…nch, r=<try> Re-enable the early otherwise branch optimization Closes rust-lang#95162. Fixes rust-lang#119014. This is the first part of rust-lang#121397. An invalid enum discriminant can come from anywhere. We have to check to see if all successors contain the discriminant statement. This should have a pass to hoist instructions. r? cjgillot

…nch, r=cjgillot Re-enable the early otherwise branch optimization Closes rust-lang#95162. Fixes rust-lang#119014. This is the first part of rust-lang#121397. An invalid enum discriminant can come from anywhere. We have to check to see if all successors contain the discriminant statement. This should have a pass to hoist instructions. r? cjgillot

Dylan-DPC · 2024-08-08T10:43:53Z

Closing this as it's being split into multiple prs with #122387 being the first part

DianQK · 2024-08-13T10:28:21Z

Part 2: #129047.

rustbot assigned michaelwoerister Feb 21, 2024

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 Feb 21, 2024

rustbot assigned wesleywiser and unassigned michaelwoerister Feb 21, 2024

cjgillot self-assigned this Feb 22, 2024

DianQK force-pushed the early_otherwise_branch_sound branch from 4613221 to f65946e Compare February 22, 2024 22:26

DianQK mentioned this pull request Feb 23, 2024

Suboptimal order of tests in match #117970

Open

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 Feb 29, 2024

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 Feb 29, 2024

DianQK added 8 commits March 3, 2024 21:43

Add test case for rust-lang#119014

0533c43

Resolve unsound hoisting of discriminant in EarlyOtherwiseBranch

c340ced

Re-enable the early otherwise branch optimization

26afcef

Add FileCheck for early_otherwise_branch*.rs

42926f6

Apply EarlyOtherwiseBranch to scalar values

9f14320

Apply EarlyOtherwiseBranch for all targets with the same value

e73ae2f

Apply EarlyOtherwiseBranch even if the types are different when the…

30b690b

… target values are the same

Don't modify the otherwise of the switch

8d0f254

DianQK force-pushed the early_otherwise_branch_sound branch from d5c6f73 to 8d0f254 Compare March 3, 2024 13:53

This comment has been minimized.

Sign in to view

rustbot removed S-waiting-on-perf Status: Waiting on a perf run to be completed. perf-regression Performance regression. labels Mar 4, 2024

DianQK mentioned this pull request Mar 6, 2024

Inefficient codegen for eq of enum with identical variants #113506

Open

cjgillot reviewed Mar 10, 2024

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 Mar 10, 2024

DianQK marked this pull request as draft March 11, 2024 01:26

DianQK changed the title ~~Re-enable the early otherwise branch optimization~~ [WIP] Re-enable the early otherwise branch optimization Mar 12, 2024

DianQK mentioned this pull request Mar 12, 2024

Re-enable the early otherwise branch optimization #122387

Merged

Dylan-DPC closed this Aug 8, 2024

DianQK deleted the early_otherwise_branch_sound branch August 12, 2024 01:28

apiraino removed the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Sep 5, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP] Re-enable the early otherwise branch optimization #121397

[WIP] Re-enable the early otherwise branch optimization #121397

DianQK commented Feb 21, 2024 •

edited

Loading

rustbot commented Feb 21, 2024

michaelwoerister commented Feb 21, 2024

bors commented Feb 22, 2024

DianQK commented Feb 29, 2024

This comment has been minimized.

bors commented Feb 29, 2024

bors commented Feb 29, 2024

This comment has been minimized.

rust-timer commented Feb 29, 2024

bors commented Mar 3, 2024

bors commented Mar 3, 2024

This comment has been minimized.

rust-timer commented Mar 4, 2024

DianQK commented Mar 4, 2024

rustbot commented Mar 4, 2024

cjgillot left a comment

cjgillot Mar 10, 2024

DianQK Mar 11, 2024

cjgillot Mar 10, 2024

cjgillot Mar 10, 2024

cjgillot Mar 10, 2024

cjgillot Mar 10, 2024

erikdesjardins commented Mar 10, 2024

DianQK commented Mar 11, 2024

DianQK commented Mar 11, 2024

DianQK commented Mar 11, 2024

DianQK commented Mar 12, 2024

DianQK commented Mar 12, 2024

Dylan-DPC commented Aug 8, 2024

DianQK commented Aug 13, 2024

[WIP] Re-enable the early otherwise branch optimization #121397

[WIP] Re-enable the early otherwise branch optimization #121397

Conversation

DianQK commented Feb 21, 2024 • edited Loading

rustbot commented Feb 21, 2024

michaelwoerister commented Feb 21, 2024

bors commented Feb 22, 2024

DianQK commented Feb 29, 2024

This comment has been minimized.

bors commented Feb 29, 2024

bors commented Feb 29, 2024

This comment has been minimized.

rust-timer commented Feb 29, 2024

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

bors commented Mar 3, 2024

bors commented Mar 3, 2024

This comment has been minimized.

rust-timer commented Mar 4, 2024

Overall result: ❌ regressions - no action needed

DianQK commented Mar 4, 2024

rustbot commented Mar 4, 2024

cjgillot left a 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

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

erikdesjardins commented Mar 10, 2024

DianQK commented Mar 11, 2024

DianQK commented Mar 11, 2024

DianQK commented Mar 11, 2024

DianQK commented Mar 12, 2024

DianQK commented Mar 12, 2024

Dylan-DPC commented Aug 8, 2024

DianQK commented Aug 13, 2024

DianQK commented Feb 21, 2024 •

edited

Loading