linux-kernel - Re: [PATCH 2/2] bitops: rotate: Add riscv implementation using Zbb extension

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20250702123043.1460-1-cp0613@linux.alibaba.com>
Date: Wed,  2 Jul 2025 20:30:43 +0800
From: cp0613@...ux.alibaba.com
To: yury.norov@...il.com
Cc: alex@...ti.fr,
	aou@...s.berkeley.edu,
	arnd@...db.de,
	cp0613@...ux.alibaba.com,
	linux-arch@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	linux-riscv@...ts.infradead.org,
	linux@...musvillemoes.dk,
	palmer@...belt.com,
	paul.walmsley@...ive.com
Subject: Re: [PATCH 2/2] bitops: rotate: Add riscv implementation using Zbb extension

On Tue, 1 Jul 2025 14:32:14 -0400, yury.norov@...il.com wrote:

> > static inline u8 variable_rol8(u8 word, unsigned int shift)
> 
> Now that it has assembler inlines, would it help to add the __pure
> qualifier?

Even if the same parameters are passed in, after comparing the disassembly
with and without the __pure qualifier, it is found that it does not help.

> Please do the following:
> 
> 1. Drop the generic_ror8() and keep only ror/l8()
> 2. Add ror/l16, 34 and 64 tests.
> 3. Adjust the 'loop' so that each subtest will take 1-10 ms on your hw.
> 4. Name the function like test_rorl(), and put it next to the test_fns()
>    in lib/test_bitops.c. Refer the 0a2c6664e56f0 for implementation.
> 5. Prepend the series with the benchmark patch, just as with const-eval
>    one, and report before/after for your series. 
> 
> > > Please find attached a test for compile-time ror/rol evaluation.
> > > Please consider prepending your series with it.
> > 
> > Okay, I'll bring it to the next series.
> 
> Still waiting for bloat-o-meter results.

Ok, in the next series I will provide a patch for the benchmark, and then
prepend the const-eval and benchmark patches, and also provide the results
of the bloat-o-meter.

Thank you very much,
Pei