JIT: boost inlining when callee unboxes an arg #110596
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Especially so if the caller is passing an exact type. This may lead to the JIT being able to stack allocate the box and promote the underlying payload. Fixes dotnet#104479
dotnet-issue-labeler
bot
added
the
area-CodeGen-coreclr
|
On the example from #104479: ; Method Program:<<Main>$>g__BoxTest|0_0():int (FullOpts)
G_M15828_IG01: ;; offset=0x0000
sub rsp, 40
;; size=4 bbWeight=1 PerfScore 0.25
G_M15828_IG02: ;; offset=0x0004
mov ecx, 42
call [System.Number:Int32ToDecStr(int):System.String]
mov rdx, rax
mov rcx, 0x7FFC28EF66C0 ; Program:<<Main>$>g__Consume|0_2[System.String](System.String)
call [Program:<<Main>$>g__Consume|0_2[System.__Canon](System.__Canon)]
mov eax, 42
;; size=35 bbWeight=1 PerfScore 7.00
G_M15828_IG03: ;; offset=0x0027
add rsp, 40
ret
;; size=5 bbWeight=1 PerfScore 1.25
; Total bytes of code: 4SPMI sees some diffs and quite a few missed contexts. MihuBot did not see many diffs, however: So we may need a bespoke SPMI to really evaluate the imapact here. |
|
@hez2010 FYI -- seems quite easy to blow code up this way, but perhaps worth considering. |
|
@EgorBo FYI |
EgorBo
reviewed
Feb 7, 2025
| @@ -1716,6 +1724,30 @@ double ExtendedDefaultPolicy::DetermineMultiplier() | |||
| JITDUMP("\nPrejit root candidate has arg that feeds a conditional. Multiplier increased to %g.", multiplier); | |||
| } | |||
|
|
|||
| if (m_ArgUnboxExact > 0) | |||
There was a problem hiding this comment.
@AndyAyersMS I think you can also rely on existing m_ArgIsBoxedAtCallsite here, so if an arg is boxed on the callsite and unboxed in the callee - it should be quite profitable.
UPD: although, probably not - m_ArgIsBoxedAtCallsite alone is not telling which arg exactly is boxed 🙂
EgorBo
approved these changes
Feb 7, 2025
|
At least from MihuBot I only see more inlining, and no cases where we removed allocations. But might as well see if the lab spots anything. |
This was referenced Feb 9, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Especially so if the caller is passing an exact type. This may lead to the JIT being able to stack allocate the box and promote the underlying payload.
Fixes #104479