Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Refactor AdvSimd version of DecodeFromUTF8#101620

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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Merged

Conversation

@SwapnilGaikwad
Copy link
Contributor

@ghostghost added the needs-area-labelAn area label is needed to ensure this gets routed to the appropriate area owners labelApr 26, 2024
@dotnet-policy-servicedotnet-policy-servicebot added the community-contributionIndicates that the PR has been added by a community member labelApr 26, 2024
@SwapnilGaikwad
Copy link
ContributorAuthor

@a74nh@kunalspathak @dotnet/arm64-contrib

@SwapnilGaikwadSwapnilGaikwad marked this pull request as ready for reviewApril 26, 2024 18:33
@SwapnilGaikwad
Copy link
ContributorAuthor

There is no notable performance difference on a V1 and N1 system for this patch.
There is reordering of assembly sequence with the newer version having an instruction less.

Assembly sequence for DecodeFromUtf8
b.cc0xffffa8338394  // b.lo, b.ul, b.lastldrq8, 0xffffa83388c0strq8, [x29, #96]ldrq9, 0xffffa83388d0strq9, [x29, #80]ldrq10, 0xffffa83388e0strq10, [x29, #64]ldrq11, 0xffffa83388f0strq11, [x29, #48]ldrq12, 0xffffa8338900strq12, [x29, #32]ldrq13, 0xffffa8338910strq13, [x29, #16]movx6, x27strx28, [x29, #272]ldrq14, 0xffffa8338920strq14, [x29, #256]strw4, [x29, #344]movw2, w4strx6, [x29, #280]movx0, x6movx1, x27adrpx11, 0xffffa8e3a000addx11, x11, #0x918movv14.d[0], v8.d[1]movv10.d[0], v9.d[1]ldrx13, [x11]blrx13ldrx6, [x29, #280]ld4{v16.16b-v19.16b}, [x6]stpq16, q17, [x29, #192]stpq18, q19, [x29, #224]ldpq16, q17, [x29, #192]ldpq18, q19, [x29, #224]mvniv20.4s, #0x0mvniv21.4s, #0x0movv8.d[1], v14.d[0]movv9.d[1], v10.d[0]movv22.16b, v8.16bmovv23.16b, v9.16btblv20.16b, {v20.16b-v23.16b}, v16.16bmvniv21.4s, #0x0mvniv22.4s, #0x0movv23.16b, v21.16bmovv24.16b, v22.16bmovv25.16b, v8.16bmovv26.16b, v9.16btblv21.16b, {v23.16b-v26.16b}, v17.16bmvniv22.4s, #0x0mvniv23.4s, #0x0movv24.16b, v22.16bmovv25.16b, v23.16bmovv26.16b, v8.16bmovv27.16b, v9.16btblv22.16b, {v24.16b-v27.16b}, v18.16bmvniv23.4s, #0x0mvniv24.4s, #0x0movv25.16b, v23.16bmovv26.16b, v24.16bmovv27.16b, v8.16bmovv28.16b, v9.16btblv23.16b, {v25.16b-v28.16b}, v19.16bldrq24, [x29, #256]uqsubv16.16b, v16.16b, v24.16buqsubv17.16b, v17.16b, v24.16buqsubv18.16b, v18.16b, v24.16buqsubv19.16b, v19.16b, v24.16bldpq26, q25, [x29, #48]ldpq28, q27, [x29, #16]tbxv16.16b, {v25.16b-v28.16b}, v16.16btbxv17.16b, {v25.16b-v28.16b}, v17.16btbxv18.16b, {v25.16b-v28.16b}, v18.16btbxv19.16b, {v25.16b-v28.16b}, v19.16borrv16.16b, v20.16b, v16.16borrv17.16b, v21.16b, v17.16borrv18.16b, v22.16b, v18.16borrv19.16b, v23.16b, v19.16bcmhiv20.16b, v16.16b, v24.16bcmhiv21.16b, v17.16b, v24.16borrv20.16b, v20.16b, v21.16bcmhiv21.16b, v18.16b, v24.16borrv20.16b, v20.16b, v21.16bcmhiv21.16b, v19.16b, v24.16borrv20.16b, v20.16b, v21.16bumaxpv20.4s, v20.4s, v20.4smovx2, v20.d[0]cmpx2, #0x0b.ne0xffffa833836c  // b.anyshlv16.16b, v16.16b, #2ushrv20.16b, v17.16b, #4orrv10.16b, v16.16b, v20.16bshlv16.16b, v17.16b, #4ushrv17.16b, v18.16b, #2orrv11.16b, v16.16b, v17.16bshlv16.16b, v18.16b, #6orrv12.16b, v16.16b, v19.16bmovw2, w19ldrx0, [x29, #272]movx1, x28adrpx11, 0xffffa8e3a000addx11, x11, #0x920movv13.d[0], v10.d[1]movv8.d[0], v11.d[1]movv9.d[0], v12.d[1]ldrx3, [x11]blrx3movv10.d[1], v13.d[0]movv11.d[1], v8.d[0]movv12.d[1], v9.d[0]ldrx7, [x29, #272]st3{v10.16b-v12.16b}, [x7]ldrx6, [x29, #280]addx6, x6, #0x40addx7, x7, #0x30ldrx3, [x29, #288]cmpx6, x3strx7, [x29, #272]strx3, [x29, #288]ldpq9, q8, [x29, #80]b.ls0xffffa8338568  // b.plaststrx6, [x29, #280]ldrx6, [x29, #280]movx4, x6ldrx7, [x29, #272]movx5, x7ldrx6, [x29, #312]cmpx4, x6b.eq0xffffa833875c

@buyaa-nbuyaa-n added area-System.Buffers and removed needs-area-labelAn area label is needed to ensure this gets routed to the appropriate area owners labelsMay 7, 2024
@dotnet-policy-service
Copy link
Contributor

Tagging subscribers to this area: @dotnet/area-system-buffers
See info inarea-owners.md if you want to be subscribed.

Copy link
Contributor

@kunalspathakkunalspathak left a comment

Choose a reason for hiding this comment

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

Added some questions/comments.

SwapnilGaikwad reacted with thumbs up emoji
Copy link
Contributor

@kunalspathakkunalspathak left a comment

Choose a reason for hiding this comment

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

LGTM. Thanks!

SwapnilGaikwad reacted with hooray emoji
@kunalspathakkunalspathak merged commit7037516 intodotnet:mainMay 9, 2024
@SwapnilGaikwadSwapnilGaikwad deleted the github-refactor-Base64Decode branchMay 10, 2024 09:02
Ruihan-Yin pushed a commit to Ruihan-Yin/runtime that referenced this pull requestMay 30, 2024
* Refactor AdvSimd version of DecodeFromUTF8* Refactor look-up table for readability* Fix the comments
@github-actionsgithub-actionsbot locked and limited conversation to collaboratorsJun 10, 2024
Sign up for freeto subscribe to this conversation on GitHub. Already have an account?Sign in.

Reviewers

@tannergoodingtannergoodingAwaiting requested review from tannergooding

1 more reviewer

@kunalspathakkunalspathakkunalspathak approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

area-System.Bufferscommunity-contributionIndicates that the PR has been added by a community member

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@SwapnilGaikwad@kunalspathak@buyaa-n
[8]ページ先頭
©2009-2025 Movatter.jp