[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAHmME9riDsWbfQe898EAGPCU1k-pasnVV=fXu_JWKdB9AJru5Q@mail.gmail.com>
Date: Wed, 4 Sep 2024 18:37:26 +0200
From: "Jason A. Donenfeld" <Jason@...c4.com>
To: Christophe Leroy <christophe.leroy@...roup.eu>
Cc: Adhemerval Zanella <adhemerval.zanella@...aro.org>, Xi Ruoyao <xry111@...111.site>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>
Subject: Re: Profiling of vdso_test_random
On Wed, Sep 4, 2024 at 1:41 PM Christophe Leroy
<christophe.leroy@...roup.eu> wrote:
>
> Hi,
>
> I'm done a 'perf record' on vdso_test_random reduced to vdso test only,
> and I get the following function usage profile.
>
> Do you see the same type of percentage on your platforms ?
>
> I would have expected most of the time to be spent in
> __arch_chacha20_blocks_nostack() but that's in fact not the case.
>
> # Samples: 61K of event 'task-clock:ppp'
> # Event count (approx.): 15463500000
> #
> # Overhead Command Shared Object Symbol
>
> # ........ ............... ...................
> ....................................
> #
> 57.74% vdso_test_getra [vdso] [.] __c_kernel_getrandom
Can you run `perf top` to see which instructions are hot? There are
some smp barriers in there which might be significant.
Powered by blists - more mailing lists