-
Notifications
You must be signed in to change notification settings - Fork 89
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
Is there a benchmark for vector crypto extension? #375
Comments
I remember SiFive doing this test |
There are a very wide range of implementation points for vector crypto, with small to wide datapaths and pipelining options. I don’t know if any implementers have published any benchmarks on their proprietary designs. From an instruction count PoV it provides a very substantial savings, but I don’t know where that is recorded, either. Perhaps someone else in the task group can find those.
G. Richard Newell
Assoc. Technical Fellow, FPGA Business Unit, Microchip Technology
(408) 643-6146 (office), (408) 882-4785 (mobile), +1 (925) 478-7258 (Skype, not monitored)
PGP: (2009 DSA-1024, ELG-4096) B751 FC13 8B4E 49DA 2270 35A2 20E4 E66A D0D0 2E34
(2016 SSA-4096, RSA-4096) 65F5 CCD6 23B3 BD3D CEDE AB58 171F F4DE E7D0 3ECA
From: ware ***@***.***>
Sent: Sunday, November 5, 2023 7:13 PM
To: riscv/riscv-crypto ***@***.***>
Cc: Subscribed ***@***.***>
Subject: Re: [riscv/riscv-crypto] Is there a benchmark for vector crypto extension? (Issue #375)
EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
I remember SiFive doing this test
—
Reply to this email directly, view it on GitHub<#375 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AKTRL7IKG6CVBN7Y3AVEJETYDBIT7AVCNFSM6AAAAAA662NSPKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOJUGAYTSMJRGA>.
You are receiving this because you are subscribed to this thread.Message ID: ***@***.******@***.***>>
|
OpenSSL has implementations for both RISC-V scalar and vector instructions. Those could be used for "apples to apples" comparisons between implementations. Note that, due to implementation-specific decisions (e.g. instruction latencies, vector width) this code may not be optimal for any particular implementation, requiring tuning. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The current repository has a scalar version of the benchmark. Is there a vector version of the benchmark in the community or a plan for it? If there is any, we are willing to participate, and thank you very much.
The text was updated successfully, but these errors were encountered: