rustc_codegen_ssa: Fix for codegen_get_discr

[mikebenfield]

When doing the optimized implementation of getting the discriminant, the arithmetic needs to be done in the tag type so wrapping behavior works correctly.

Fixes #104519

[rustbot]

r? @lcnr

(rustbot has picked a reviewer for you, use r? to override)

[mikebenfield]

If desired I can provide a more thorough explanation of what the problem was, either here in a PR comment, in the commit message, or in code comments.

[lcnr]

r? compiler

[pnkfelix]

@bors r+

[bors]

📌 Commit 901f8d81b84bdf5dadd41896ed8f663ab17cc039 has been approved by pnkfelix

It is now in the queue for this repository.

[bors]

🌲 The tree is currently closed for pull requests below priority 1. This pull request will be tested once the tree is reopened.

[mikebenfield]

Alright, obviously I needed to update the codegen test too.

Interestingly I think it's likely moving the arithmetic before the cast may result in better code, as maybe this LLVM issue won't matter.

[mikebenfield]

I don't understand this failure. The test passes on my system, and in the error message here the "possible intended match" appears to be identical to the text it's supposed to match.

[mikebenfield]

I realized the difference in the test; apparently the call is not always a tail call.

[djkoloski]

Is this ready to merge?

[klensy]

There was pushes after r+, so need to reapprove this.

[djkoloski]

Seems like the labels are incorrect then, since this has S-waiting-on-bors

[mikebenfield]

This should be ready, yes.

[jackh726]

@bors r=pnkfelix

[bors]

📌 Commit 31c0645 has been approved by pnkfelix

It is now in the queue for this repository.

[bors]

⌛ Testing commit 31c0645 with merge 53e4b9d...

[bors]

☀️ Test successful - checks-actions
Approved by: pnkfelix
Pushing 53e4b9d to master...

[rust-timer]

Finished benchmarking commit (53e4b9d): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -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
Regressions ❌ (secondary)	0.4%	[0.3%, 0.5%]	7
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

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)	2.2%	[2.2%, 2.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.8%	[-1.8%, -1.8%]	1
All ❌✅ (primary)	-	-	0

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)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.5%	[-2.9%, -2.0%]	2
All ❌✅ (primary)	-	-	0