Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <CAHk-=wibrw+PnBiQbkGy+5p4GpkPwmmodw-beODikL-tiz0dFQ@mail.gmail.com>
 <20210429100508.18502-1-arek_koz@o2.pl>
In-Reply-To: <20210429100508.18502-1-arek_koz@o2.pl>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Thu, 29 Apr 2021 09:36:22 -0700
Message-ID: <CAHk-=wgYiKOYTtU6DifULbj0tmFLJf2Va5ScZW0dCgWi8=-c1A@mail.gmail.com>
Subject: Re: [PATCH v2] proc: Use seq_read_iter for /proc/*/maps
To:     "Arkadiusz Kozdra (Arusekk)" <arek_koz@o2.pl>
Cc:     Christoph Hellwig <hch@lst.de>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Alexey Dobriyan <adobriyan@gmail.com>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Apr 29, 2021 at 3:04 AM Arkadiusz Kozdra (Arusekk)
<arek_koz@o2.pl> wrote:
>
> Since seq_read_iter looks mature enough to be used for /proc/<pid>/maps,
> re-allow applications to perform zero-copy data forwarding from it.

I'd really like to hear what the programs are, and what the
performance difference is.

Because I'm surprised that the advantages of splice would really be
noticeable. I don't _dispute_ it, but I really would like this to be
actually _documented_, not just "Some executable-inspecting tools".

What tools (so that if it causes issues later, we have that
knowledge), and what are the performance numbers?

             Linus