Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Use IntegralRange in fgOptimizeRelationalComparisonWithCasts #69247

Merged

Conversation

jakobbotsch
Copy link
Member

@jakobbotsch jakobbotsch commented May 12, 2022

@dotnet-issue-labeler dotnet-issue-labeler bot added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label May 12, 2022
@ghost ghost assigned jakobbotsch May 12, 2022
@ghost
Copy link

ghost commented May 12, 2022

Tagging subscribers to this area: @JulieLeeMSFT
See info in area-owners.md if you want to be subscribed.

Issue Details

Fix #69200

Author: jakobbotsch
Assignees: -
Labels:

area-CodeGen-coreclr

Milestone: -

@jakobbotsch

This comment was marked as outdated.

@jakobbotsch jakobbotsch marked this pull request as draft May 12, 2022 14:25
@jakobbotsch
Copy link
Member Author

Diff for example in #69200:

        mov      qword ptr [rsp+28H], rax
 						;; size=11 bbWeight=1    PerfScore 1.50
 G_M28380_IG02:
-       mov      eax, edx
-       mov      r8d, dword ptr [rcx+8]
-       cmp      rax, r8
+       cmp      edx, dword ptr [rcx+8]
        ja       SHORT G_M28380_IG04
        mov      rcx, bword ptr [rcx]
        mov      bword ptr [rsp+28H], rcx
@@ -43,7 +41,7 @@ G_M28380_IG02:
        lea      rcx, bword ptr [rsp+28H]
        call     [Program:DoIt(ReadOnlySpan`1)]
        nop      
-						;; size=35 bbWeight=1    PerfScore 11.25
+						;; size=29 bbWeight=1    PerfScore 11.75
 G_M28380_IG03:
        add      rsp, 56
        ret      
@@ -53,6 +51,6 @@ G_M28380_IG04:
        int3     
 						;; size=7 bbWeight=0    PerfScore 0.00
 
-; Total bytes of code 58, prolog size 11, PerfScore 19.80, instruction count 17, allocated bytes for code 58 (MethodHash=f8b59123) for method Program:M(ReadOnlySpan`1,int)
+; Total bytes of code 52, prolog size 11, PerfScore 19.70, instruction count 15, allocated bytes for code 52 (MethodHash=f8b59123) for method Program:M(ReadOnlySpan`1,int)
 ; ============================================================

@jakobbotsch
Copy link
Member Author

/azp run Fuzzlyn

@azure-pipelines
Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@jakobbotsch jakobbotsch marked this pull request as ready for review May 16, 2022 20:13
@jakobbotsch
Copy link
Member Author

cc @dotnet/jit-contrib @GrabYourPitchforks

Copy link
Contributor

@SingleAccretion SingleAccretion left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks great (modulo nits)!

src/coreclr/jit/morph.cpp Outdated Show resolved Hide resolved
src/coreclr/jit/morph.cpp Outdated Show resolved Hide resolved
@jakobbotsch jakobbotsch requested review from TIHan and EgorBo May 19, 2022 19:08
@jakobbotsch jakobbotsch merged commit 534cfe3 into dotnet:main May 23, 2022
@jakobbotsch jakobbotsch deleted the better-rel-comparison-with-casts-opts branch May 23, 2022 17:00
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI
Projects
None yet
Development

Successfully merging this pull request may close these issues.

JIT perf: Span.Slice(0, ...) emits unnecessary 64-bit extensions
3 participants