[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250510060219.GC505731@sol>
Date: Fri, 9 May 2025 23:02:19 -0700
From: Eric Biggers <ebiggers@...nel.org>
To: Herbert Xu <herbert@...dor.apana.org.au>
Cc: Venkat Rao Bagalkote <venkat88@...ux.ibm.com>,
Thorsten Leemhuis <linux@...mhuis.info>,
Linux Crypto Mailing List <linux-crypto@...r.kernel.org>,
LKML <linux-kernel@...r.kernel.org>,
Linux Next Mailing List <linux-next@...r.kernel.org>,
Madhavan Srinivasan <maddy@...ux.ibm.com>,
Stephen Rothwell <sfr@...b.auug.org.au>,
Danny Tsen <dtsen@...ux.ibm.com>, linuxppc-dev@...ts.ozlabs.org
Subject: Re: [v2 PATCH] crypto: powerpc/poly1305 - Add poly1305_emit_arch
wrapper
On Sat, May 10, 2025 at 01:50:02PM +0800, Herbert Xu wrote:
> On Sat, May 10, 2025 at 01:49:13PM +0800, Herbert Xu wrote:
> >
> > Did I do something wrong?
>
> OK perhaps it's just that the qemu emulation being slow.
Yes, non-native QEMU usually isn't any good for benchmarking the
architecture-optimized code, due to the instructions it uses having to be
emulated. Just to give another random example, in (non-native) QEMU the RISC-V
CRC code is much slower than the generic CRC code. But when run on an actual
RISC-V processor it's much faster.
- Eric
Powered by blists - more mailing lists